Category Archives: Robotics

Oil-and-Gas-Robotics-Market

Latest research on Global Oil and Gas Robotics Market report covers forecast and analysis on a worldwide, regional and country level. The study provides historical information of 2016-2021 together with a forecast from 2021 to 2026 supported by both volume and revenue (USD million). The entire study covers the key drivers and restraints for the Oil and Gas Robotics market. this report included a special section on the Impact of COVID19. Also, Oil and Gas Robotics Market (By major Key Players, By Types, By Applications, and Leading Regions) Segments outlook, Business assessment, Competition scenario and Trends .The report also gives 360-degree overview of the competitive landscape of the industries.

Moreover, it offers highly accurate estimations on the CAGR, market share, and market size of key regions and countries. Players can use this study to explore untapped Oil and Gas Robotics markets to extend their reach and create sales opportunities.

Some of the key manufacturers operating in this market include: iRobot Corporation, ABB Ltd, Fanuc Corporation, Delaval Group, Lely Group, Kuka AG, Yaskawa Electric Corporation and More

Download Free PDF Sample Copy of the Report(with covid 19 Impact Analysis): https://www.globmarketreports.com/request-sample/90073

Our Research Analyst implemented a Free PDF Sample Report copy as per your Research Requirement, also including impact analysisofCOVID-19 on Oil and Gas Robotics Market Size

Oil and Gas Robotics market competitive landscape offers data information and details by companies. Its provides a complete analysis and precise statistics on revenue by the major players participants for the period 2021-2026. The report also illustrates minute details in the Oil and Gas Robotics market governing micro and macroeconomic factors that seem to have a dominant and long-term impact, directing the course of popular trends in the global Oil and Gas Robotics market.

Product Type Coverage (Market Size & Forecast, Major Company of Product Type etc.): Remotely Operated Vehicles Autonomous Underwater Vehicles Uavs & Unmanned Ground VehiclesApplication Coverage (Market Size & Forecast, Different Demand Market by Region, Main Consumer Profile etc.): Inspection Monitoring & Surveillance Others

Regions Covered in the Global Oil and Gas Robotics Market:1. South America Oil and Gas Robotics Market Covers Colombia, Brazil, and Argentina.2. North America Oil and Gas Robotics Market Covers Canada, United States, and Mexico.3. Europe Oil and Gas Robotics Market Covers UK, France, Italy, Germany, and Russia.4. The Middle East and Africa Oil and Gas Robotics Market Covers UAE, Saudi Arabia, Egypt, Nigeria, and South Africa.5. Asia Pacific Oil and Gas Robotics Market Covers Korea, Japan, China, Southeast Asia, and India.Years Considered to Estimate the Market Size:History Year: 2015-2021Base Year: 2021Estimated Year: 2021Forecast Year: 2021-2026

Get Chance of up to 50% Extra [emailprotected]: https://www.globmarketreports.com/request-discount/90073

Reasons to buy:

Some Major TOC Points:

For More Information with including full TOC: https://www.globmarketreports.com/industry-reports/90073/Oil-and-Gas-Robotics-market

Key highlights of the Oil and Gas Robotics Market report: Growth rate Renumeration prediction Consumption graph Market concentration ratio Secondary industry competitors Competitive structure Major restraints Market drivers Regional bifurcation Competitive hierarchy Current market tendencies Market concentration analysisCustomization of the Report: Glob Market Reports provides customization of reports as per your need. This report can be personalized to meet your requirements. Get in touch with our sales team, who will guarantee you to get a report that suits your necessities.

Get Customization of the [emailprotected]: https://www.globmarketreports.com/request-customization/90073

Contact Us:Glob Market Reports17224 S. Figueroa Street,Gardena, California (CA) 90248,United StatesCall:+1 915 229 3004(U.S)+44 7452 242832(U.K)Website: http://www.globmarketreports.com

For More Reports Click Here:

1) http://marketsresearchreport.com/2021/03/02/modular-uninterruptible-power-supply-market-2021-global-industry-size-share-business-growth-revenueschneider-electric-huawei-abb-eaton-emerson-legrand-etc/2) http://marketsresearchreport.com/2021/03/02/commercial-combi-ovens-market-size-key-players-analysis-competitive-scenario-opportunities-development-status-2021-2026alto-shaam-middleby-retigo-henny-penny-itw-rational-etc/3) http://marketsresearchreport.com/2021/03/02/corporate-car-sharing-market-increasing-demand-with-industry-professionals-ubeeqo-ald-automotive-arval-sixt-fleetster-drivenow-etc/4) http://marketsresearchreport.com/2021/03/02/high-speed-doors-market-will-reflect-significant-growth-prospects-of-us-mn-during-2021-2026-with-major-key-player-hormann-rite-hite-asi-doors-rytec-assa-abloy-chase-doors-etc/5) http://marketsresearchreport.com/2021/03/02/special-mission-aircraft-market-increasing-demand-with-industry-professionals-boeing-company-bombardier-aerospace-lockheed-martin-corporation-northrop-grumman-corporation-etc/6) http://marketsresearchreport.com/2021/03/02/global-fruit-ripening-gas-market-to-witness-huge-gains-over-2021-2026saudi-basic-industries-corporation-dow-exxon-mobil-shell-sinopec-chevron-phillips-etc/

Read the original post:

Oil and Gas Robotics Market Will Increase Demand In Forecast By 2026 |iRobot Corporation, ABB Ltd, Fanuc Corporation, Delaval Group, Lely Group, etc ...

Microscopic swimming robots that could navigate through the body to perform medical tasks such as delivery of targeted cancer therapies or surgeries are currently in development. In a study published March 24 in Science Robotics, scientists made magnetically controlled microrobots based on neutrophils, a type of white blood cell. In mice, these so-called neutrobots penetrated the blood-brain barrier (BBB) to deliver drugs to brain cancer cells.

This is a very cool idea, says Liangfang Zhang, a nanoengineer and bioengineer at the University of California, San Diego, who was not involved with the study. I would say this paper is still an early proof-of-concept study, but I think that the overall concept is novel. Its interesting because its new thinking about how to send cargo to the brain.

A major hurdle in treating neurological diseases is getting drugs past the BBB, a highly selective boundary that denies most substances admission to the brain. But certain white blood cells are granted special access to deal with infections and inflammation, making them good trojan horses for getting drugs past this blockade. In previous studies, researchers have loaded brain cancer drugs into neutrophils and macrophages, which have a natural ability to scout out cancer because they swim toward higher concentrations of inflammatory chemicals released by diseased tissue.

But prior iterations of drug-ferrying immune cells have failed to completely treat mouse brain tumors, likely due in part to slow migration to the disease site. To improve speed and control, researchers have endowed microrobots based on sperm, bacteria, or red blood cells with magnetic material to externally guide them with magnetic fields, says Zhiguang Wu, a bioengineer at the Harbin Institute of Technology in China and a coauthor of the new study.

To treat glioma, a type of brain cancer, in mice, Wu and his colleagues designed neutrophil-based microrobotsneutrobotsthat could be controlled with a magnetic field. First, the team made nanoparticles from a gel embedded with magnetic iron oxide beads and the widely used cancer drug paclitaxel. Next, the nanoparticles were enrobed in E. colibacterial membrane. Disguised as harmful bacteria, the nanoparticles were engulfed by mouse neutrophils in vitro much more readily than bare nanoparticles. The bacterial cloak also prevented the premature leakage of drugs and made the particles less toxic to the neutrophils, the researchers found.

A transmission electron microscopy image of a single neutrobot. The yellow arrow indicates a cluster of nanoparticles containing iron oxide and paclitaxel, each enclosed by an E. coli membrane. The scale bar is 2 m.

The team tested the neutrobots navigation and drug-delivery capabilities in vitro. Under the control of a rotating magnetic field, the neutrobots reached a speed of 16.4 m per second, about 50 times faster than the speed of natural neutrophils. By monitoring the neutrobots via a microscope, the researchers could direct them to move in complex orientations on an artificial substrate.

To evaluate the neutrobots inflammation-seeking ability, the researchers placed them in a gel with a concentration gradient of an inflammatory factor. The neutrobots migrated toward higher concentrations of the chemical at a speed on par with natural neutrophils. And in a model BBB, neutrobots penetrated mouse cells grown on a membrane to access glioma cells and released their drug payload upon exposure to inflammation signals.

Finally, the researchers tested whether the bots could treat brain cancer in mice. First, they injected glioma cells into mouse brains. After 10 days, they performed surgery on some of the mice to remove a portion of the tumor in order to boost neutrophil-attracting inflammatory signals. The researchers injected neutrobots into the tails of all of the mice, and in a subset of mice, they used a rotating magnetic field to direct the neutrobots toward the brain. Using magnetic resonance imaging (MRI), the team found that more neutrobots accumulated around gliomas in mice treated with both surgery and the magnetic field compared with mice that werent exposed to the magnetic field, didnt undergo surgery, or received neither. The doubly treated mice also survived longerevidence that the two interventions complemented one another. Transmission electron microscopy confirmed that neutrobots penetrated the BBB and entered glioma tissue.

All of the neutrobot-treated mice survived longer compared with animals treated with an injection of just saline or paclitaxel, indicating that neutrobots could still deliver drugs across the BBB in response to a weak inflammatory signal or a strong inflammatory signal without magnetic propulsion.

According to Zhang, the individual components of the studythe use of immune cells as drug carriers, magnetically controlled nanoparticles, and bacterial membranes as cloaksare not new. But what they did is integrate these common individual components together and assembled them into a new system, he says. They [developed] a very unique functionalitythat is, the long-range control of neutrophils.

Mariana Medina-Snchez, a bioengineer at the Leibniz Institute for Solid State and Materials Research Dresden in Germany who did not contribute to the research, says the study is valuable because it demonstrates effective treatment of tumors in vivo, a goal of many researchers in the field. [The study] is complete, its systematic, and there is strong evidence that what theyve developed is functioning, she says.

If you know the amount of the drug that you load per microrobot, you can control the drug dose by swarming these microrobots in a controlled way.

Mariana Medina-Snchez, Leibniz Institute for Solid State and Materials Research Dresden

But before microrobots can be used to treat cancer in people, there are still a number of challenges that need to be overcome. One of these is improving the percentage of microrobots that make it to the tumor. They had an accumulation of these neutrophil-based microrobots of about eleven percent in the disease site [in vivo]. So what happens with the others? says Medina-Snchez. Microrobots could accumulate in other organs or regions of the body, and the long-term side effects are unknown, she says. But this happens for every type of microrobot, not just for this particular work. This is a challenge for everyone to [overcome].

Once the microrobots arrive at the disease site, another hurdle is making sure they deliver enough of the drug. You need to increase the overall drug payload inside, and you also need to control premature drug release, says Zhang. It takes time for the neutrophil to get to the destination. You dont want to them to dump all the payload before they get to the destination.

Because a single microrobot cant carry enough medication to treat a disease, researchers are also trying to understand how they move as swarmssimilar to the collective movements of groups of ants, fish, or birds. If you know the amount of the drug that you load per microrobot, you can control the drug dose by swarming these microrobots in a controlled way, says Medina-Snchez. So this is one of the challenges: how to transport multiple [microrobots] in a controlled manner and deliver them to a target location. Wu and his colleagues found that neutrobots formed chains of four in vitro, and these swarms swam about five times faster than individual bots did. But according to Medina-Snchez, other microrobot researchers are aiming for swarms of hundreds, thousands, or even millions. It depends on the target and location, she says. You may need just a few or millions of them.

Its not clear how the neutrobots swarmed in mice because current imaging techniques arent good enough to track individual or small chains of microrobots in real time at high enough resolution in vivoanother challenge for precise navigation of these tiny drug couriers in humans.

H. Zhang et al., Dual-responsive biohybrid neutrobots for active target delivery,Sci Robot,doi:10.1126/scirobotics.aaz9519, 2021.

Read the original:

Microscopic Robots Deliver Drugs to the Brain - The Scientist

A team of researchers from Johns Hopkins discussed how the COVID-19 crisis has accelerated new advances in automation, while also unveiling bottlenecks in the rollout of robotic systems in health care settings.

They argue that advances in human-robot interactions like enhancing robots' capabilities for touch, feeling, and decision-making will decide whether tomorrow's robots will help hospitals keep their lead on the encroaching pressure of the next global pandemic, according to an article shared on Nature Machine Intelligence.

The research team observed three ways robots have significantly advanced both patient care and provider safety amid the COVID-19 crisis. Robotic technology minimized contact between patients who had contracted the virus and health care workers, while also reducing the need for PPE, and also freed up time for health care workers to devote more attention to other crucial tasks. But the team of researchers looked forward, anticipating how advances could be leveraged to further adapt and enhance the reliability of robots for similar health calamities of the future.

Involved in the discussion were commentary authors Russel Taylor and Axel Krieger of the Whiting School of Engineering, along with Director Brian Garibaldi of the Johns Hopkins Biocontainment Unit.

"You go into a pandemic with the robots you have, not the robots you wish you had," explained Taylor, in the Nature Machine Intelligence article. "We can't build a fleet of robots for an emergency and put them in a warehouse. Not only is that not economically viable, but by the time you need them, they could be obsolete." This means advances in robotics and automated services surrounding health care require new "core capabilities into deployed systems that can be easily adapted for the challenges of the moment."

When the pandemic hit hospitals, there were already robots capable of delivering meals and taking a patient's temperature, explained Taylor. "Now we are talking about much more sophisticated systems that can do serious cleaning, that can perform nursing tasks, that can do many things beyond just delivering supplies." But these new capabilities create serious engineering challenges.

One of the major challenges revolves around deployability and how rapidly non-expert users can adapt and customize the robot for specialized scenarios. "For example, our ICU ventilator robot was designed for one kind of ventilator that pushes buttons," said Taylor. "But some ventilators have knobs, so we need to be able to add a modality so that the robot can also manipulate knobs."

"Say you want one robot that can service multiple ventilators; then you'd need a mobile robot with an arm attachment, and that robot could also do plenty of other useful jobs on the hospital floor," Taylor said.

"The pandemic has shown some of the current limitations of robotic systems to robustly work and adapt in difficult, changing environments at large scales," said Krieger, toNature Machine Intelligence. The greater degree of uncertainty and chaos surrounding the unexpected in hospitals is taxing on any system, robotic or not. One strategy for overcoming this is implementing health care robots with shared autonomy, "which combines the knowledge of medical experts with the capabilities of robots."

Unlike human health care workers, robots don't need to wear fresh PPE every time they move close to an infectious patient and this "frees up valuable supplies and time for human providers," said Garibaldi, in theNature Machine Intelligence article.

However, a major point of potential improvement for robots lies in advancing their ability to execute fine motor tasks, so they can offer more comprehensive health care service, like "placing an IV, intubating the trachea, or inserting central lines," explained Garibaldi. "Other potential tasks could include basic room cleaning, phlebotomy, and ventilator and monitor management and manipulation."

However, there are some tasks for which patients will prefer human health providers for the foreseeable future. In addition to the moral support and empathy offered by a living, breathing human, there are instances when caregivers will still say "I'm not sure I can trust a robot to do that," said Taylor. "Engineers need feedback on how these systems really work in the wild."

The research team is exploring ways of enhancing ICU robots with an emphasis on "higher accuracy and higher fidelity operation of ventilators," said Krieger. Future health care robots might also carry out lung ultrasound imaging via 3D cameras and force sensors in addition to advanced autonomous surgical robotic procedures (like suturing). With a more efficient feedback loop between deployment, implementation, and development of robot and automated systems, the next generation of health care robots could eventually prove a match tothe logistical chaos of pandemics, move for move.

Read the original here:

The Next Pandemic May Be in the Hands of...Human-Like Robots? - Interesting Engineering

The research report on the Surgical Robotics and Navigation Market provides in-depth current market analysis scenario, upcoming as well as future opportunities, revenue growth, pricing and profitability by Data Bridge Market Research. The market study conducted in this report analyzes the market status, market share, growth rate, market drivers, opportunities and challenges, risks and entry barriers, sales channels, and distributors in the industry. In addition, research report offers an all-inclusive assessment of the market. Our expert analysts studied the impacts of the Covid-19 pandemic on the Surgical Robotics and Navigation Market. The global Surgical Robotics and Navigation analysis consists primarily of products that used an accurate deployment. The report further identifies the past, present, and future trends that are expected to influence the development rate of the market. Surgical Robotics and Navigation market consists of present as well as future data for the forecast period from 2021 to 2028, and also provides compounded annual growth rate (CAGR%), which is measured for regional markets and individual segment-wise.

Surgical robotics and navigation market is expected to gain market growth in the forecast period of 2020 to 2027. Data Bridge Market Research analyses the market to account to USD 4.20 billion by 2027 growing at a CAGR of 6.80% in the above-mentioned forecast period. The surgical robotics and navigation business is anticipated to develop at a strong pace due to an addition in the old aged community, acceleration in the selection of surgical navigation technology in minimally interfering operational methods, and large pervasiveness of chronic dysfunctions such as cerebrovascular diseases, strokes, andAlzheimers.

Download Free Exclusive Sample (350 Pages PDF) Report: To Know the Impact of COVID-19 on this Industry @https://www.databridgemarketresearch.com/request-a-sample/?dbmr=global-surgical-robotics-and-navigation-market

The major players of the Surgical Robotics and Navigation market are:

Surgical Robotics and Navigation Market Segmentation:

By Type (Neurosurgery Navigation, Spinal Surgical Navigation, ENT Surgical Navigation, Orthopedic Surgical Navigation, Robotic-Assisted Spinal Surgery Systems, Robotic-Assisted Neurosurgery Systems, Robotic Radiosurgery Systems, Others)

By End Use (Hospitals, Ambulatory Surgical Centers, Others)

By Application (Child, Adults, Old Aged)

Research Methodology:

The research study Surgical Robotics and Navigation market regarding size, growth, opportunities and competitive analysis has been prepared with the help of primary research, secondary research and panel review. The secondary research includes: industry publications, industry reports, trade magazines, and other publications from government and trade associations, among others. Further, the analysis done through various news articles and press releases published on reliable news journals required for every market enthusiast, policymaker, investor, and market player. The Surgical Robotics and Navigation Market is also aids in developing counter approaches for major challenges faced by the industry. A reliable report is structured with the vigilant efforts of innovative, enthusiastic, knowledgeable and experienced team of analysts, researchers, industry experts, and forecasters.

Table of Content

Chapter 1: Introduction and Scope

Chapter 2: Key Company Profiles

Chapter 3: Surgical Robotics and Navigation Market Explanations, Share and Forecast across type, application and geography

Chapter 4: Surgical Robotics and Navigation Industry Consumption by Regions

Chapter 5: Market Explanation of Asia Pacific region

Chapter 6: Market Explanations of Europe region

Chapter 7: Market Explanation of Asia Pacific region

Chapter 8: Market Explanations of North America region

Chapter 9: Market Explanations of Middle East and Africa region

Chapter 10: Key Important features of the Surgical Robotics and Navigation market

Chapter 11: Key trends of the market and the market Opportunities

Chapter 12: Industrial Chain, Sourcing Strategy and Downstream Buyers

Chapter 13: Research Findings and Conclusion

Read complete report along with TOC @https://www.databridgemarketresearch.com/toc/?dbmr=global-surgical-robotics-and-navigation-market

Global Surgical Robotics and Navigation Market Scope and Market Size

On the basis of type, the surgical robotics and navigation market is segmented into neurosurgery navigation,spinal surgicalnavigation, ENT surgical navigation, orthopedic surgical navigation, robotic-assisted spinal surgery systems, robotic-assisted neurosurgery systems, robotic radiosurgery systems, and others.

On the basis of end use, the surgical robotics and navigation market is segmented into hospitals, ambulatory surgical centers, and others.

On the basis of application, the surgical robotics and navigation market is bifurcated into child, adults and old aged.

Competitive Landscape and Surgical Robotics and NavigationMarket Share Analysis

The major players covered in the surgical robotics and navigation market report are Medtronic , Stryker, GENERAL ELECTRIC COMPANY, Brainlab, B. Braun Melsungen AG, Johnson & Johnson Services, KARL STORZ SE & Co. KG, Zimmer Biomet, Fiagon AG Medical Technologies, XION GmbH, Collin SAS, Medrobotics Corporation, Titan Medical Inc., Verb Surgical Inc, among other domestic and global players. Market share data is available for global, North America, Europe, Asia-Pacific (APAC), Middle East and Africa (MEA) and South America separately. DBMR analysts understand competitive strengths and provide competitive analysis for each competitor separately.

The key questions answered in this report:

Enquire for customization in Report @:https://www.databridgemarketresearch.com/inquire-before-buying/?dbmr=global-surgical-robotics-and-navigation-market

Surgical Robotics and Navigation Market Country Level Analysis

About Data Bridge Market Research Private Ltd:

Data Bridge Market ResearchPvt Ltdis a multinational management consulting firm with offices in India and Canada. As an innovative and neoteric market analysis and advisory company with unmatched durability level and advanced approaches. We are committed to uncover the best consumer prospects and to foster useful knowledge for your company to succeed in the market.

Data Bridge Market Research is a result of sheer wisdom and practice that was conceived and built-in Pune in the year 2015. The company came into existence from the healthcare department with far fewer employees intending to cover the whole market while providing the best class analysis. Later, the company widened its departments, as well as expands their reach by opening a new office in Gurugram location in the year 2018, where a team of highly qualified personnel joins hands for the growth of the company. Even in the tough times of COVID-19 where the Virus slowed down everything around the world, the dedicated Team of Data Bridge Market Research worked round the clock to provide quality and support to our client base, which also tells about the excellence in our sleeve.

Data Bridge Market Research has over 500 analysts working in different industries. We have catered more than 40% of the fortune 500 companies globally and have a network of more than 5000+ clientele around the globe.

Contact:

Data Bridge Market Research

US: +1 888 387 2818

UK: +44 208 089 1725

Hong Kong: +852 8192 7475

Email:[emailprotected]

Browse Related Reports:

https://fillyourarticles.com/europe-orthopedic-prosthetics-market-share-application-analysis-regional-outlook-competitive-strategies-forecast-up-to-2027/

Malaysia Elderly Care Market by Regional Outlook, Latest Trend, Share Analysis, Growth, Application and Forecast to 2027

https://articlesmaker.com/long-read-sequencing-market-size-status-growth-global-production-demand-and-business-outlook-2027/

Europe Aromatherapy Market 2021 Industry Share, Growth Opportunity, Technology Innovation and Top Manufacturers Survey by 2027

Follow this link:

Surgical Robotics and Navigation Market, Share, Application Analysis, Regional Outlook, Competitive Strategies & Forecast up to 2027 || Fiagon AG...

One company built an autonomous vehicle to help haul crops, saving work and time. Others use drones and sensors to communicate with farmers.

Image: iStock/lamyai

Farmers have long struggled with operational optimization and labor concerns. Finding enough labor to get the job done, as well as keeping workers safe is a constant struggle.

"There is an immediate need to improve efficiency and reduce costs, especially now that the pandemic has exposed just how fragile the supply chain is," said Suma Reddy, CEO of Future Acres, an agricultural robotics and artificial intelligence company. "We saw shortages in both production and more workers being put at risk when picking specialty crops on a daily basis that have really caused the industry to take a step back and re-examine how we can create greater resiliency in the food chain."

SEE: Natural language processing: A cheat sheet (TechRepublic)

One idea is to equip farms with a combination of AI and robotics that can "think through" as well as do some of the physical work of farming.

"We introduced Carry for that purpose," Reddy said. "It's an autonomous, electric agricultural robotic harvest companion to help farmers gather hand-picked crops faster and with less physical demand."

The Future Acres Carry helps transport harvested crops using AI.

Image: Future Acres

The self-driving Carry vehicle uses a combination of AI, automation and electric power to transport up to 500 pounds of crops. Reddy estimates that Carry can increase production efficiency by up to 30%, paying for the vehicle investment in 80 days.

"Our initial launch was targeted at customers at small- to medium-sized table-grape farms in the U.S. that are larger than 100 acres," Reddy said. "Grapes were the specialty crop we focused on initially, but the specialty crop market covers more than just grapes, and we believe that Carry can improve the harvesting of those types of crops as well."

Morder Intelligence estimates that the AI market in agriculture, valued at $766.41 million in 2020, will reach $2.5 billion by 2026. This is a compound annual growth rate of 21.52% between 2021 and 2026.

SEE: Smart farming: How IoT, robotics, and AI are tackling one of the biggest problems of the century (TechRepublic)

In this market, Carry is just one example of an array of autonomous technologies in agriculture that include AI, robotics and automation. Other examples are autonomous tractors and harvesters, as well as aerial drones that map fields and identify topography, soil types and moisture content from the air to provide input for prescriptive fertilizers that AI develops in order to optimize crop yields.

"In our case, we wanted to provide a robotic harvest companion that can transport up to 500 pounds of crops on all types of terrain and in all weather conditions," Reddy said. "To do this, we use machine learning and computer vision capabilities that enable the vehicle to avoid obstacles like trees and people, and to collect and apply data to further optimize precision and efficiency."

SEE: Future of farming: AI-enabled harvest robot flexes new dexterity skills (TechRepublic)

As with any technological advancement, trial-and-error proofs of concept are needed. Farming operational habits also need to be changed in order to take advantage of new technology.

What Reddy and others in the field have learned is that trialing AI and robotics in actual use cases offers the only true test of how well the technology performs. This is a universal truth for all types of AI and roboticsnot just the ones that find themselves in a farmer's field.

As a one-time Peace Corps volunteer in Africa, Reddy wanted to "build a better bridge between how we manage our resources and build a better future." Her company and others are now transforming agriculture with the help of big data, analytics and hardware, and it can't come too soon. The United Nations estimates that in 30 years, the global population will reach 9.7 billion people, and there will be a need to provide 50% more food by 2050.

Now is the time for AI and robotics solution providers to jump in.

Learn the latest news and best practices about data science, big data analytics, and artificial intelligence. Delivered Mondays

Read more from the original source:

AI and robotics are helping optimize farms to increase productivity and crop yields - TechRepublic

On March 22, 2021, two companiesDaedalean and Reliable Roboticsannounced their new partnership to build advanced navigation and situational awareness (SA) systems for commercial aircraft operations. With the new certifiable technology, onboard or remote pilots will benefit from next-generation flight automation systems.The proprietary solution enables onboard pilots and remote pilots to make faster, better-informed decisions based on the advanced sensors provided by the system.

Reliable Robotics is a leader in aircraft automation. During the last months, they demonstrated pioneering capabilities of their systems byremotely piloting a Cessna 208 Caravan from a control centre in their headquarters over 50 miles away. In 2019, the company made aviation history operating a remotely piloted Cessna 172 Skyhawk over a populated region with no one on board and subsequently demonstrated a fully automated landing of the larger Cessna 208 in 2020 on the third day of flight testing.

Daedaleans systems can now feed this information about the aircraft position relative to the terrain with its obstacles and safe landing sites and relative to other traffic, to the Reliable Robotics flight control stack, which then has at its disposal an additional layer of safety and can use it to deal with multiple contingencies such as jammed or disabled GPS, non-cooperative traffic, or emergency landing scenarios.

The end product both companies foresee is a system that can operate in airspaces as a model citizen, that enables denser economic use of the airspace, at safety levels that are an order of magnitude above todays standards.

Reliable Robotics has the most credible system for remotely piloted operations with immediate applications for cargo operators, said Luuk van Dijk, Founder and CEO of Daedalean. Our team has developed advanced machine learning that can adapt to the inherent uncertainties in airspace and increasing levels of onboard autonomy. Bringing our core competencies together was a logical next step to jointly develop a solution set that makes aircraft safer.

Both companies have been built on the principle that certification is paramount from day one, said Robert Rose, Co-founder and CEO of Reliable Robotics. Daedalean is the recognized leader when it comes to developing machine learning systems within the required regulatory framework. This is not a domain where you build something first and then figure out how to certify it later.

More here:

Daedalean reveals partnership with Reliable Robotics - sUAS News

Launching today, robotics technology player Vaarst will give offshore and marine robotics new capabilities through retrofitted artificial intelligence and autonomy.

Bristol, UK; 24 March 2021. Vaarst, a technology spin-off from leading subsea robotic and hydrographic survey company Rovco, was formally launched today with the goal of revolutionising the offshore robotics sector - leveraging intelligent data flows for smart asset management and creating an energy-efficient and more sustainable future.

Vaarst will target the energy and marine sectors through its innovative technologies, such as SubSLAM X2 - an intelligent data collection system that delivers robotic spatial awareness and live 3D point clouds to any device in the world, without costly positioning systems, thereby saving many project days. This, combined with the company's machine learning and autonomy expertise will then provide the very best in efficient data collection and AI interpretation.

The new spinout company, Vaarst, is predicting immediate 2021 revenues over 1m rising to 20m+ rapidly in the next few years.

Vaarst CEO and Founder, Brian Allen, said: "Autonomous robotics are the key to reducing the cost of offshore operations. At the same time, digitalisation of field assets is essential as the industry evolves, marrying these two concepts is needed to realise the real benefit of modern tech. It's the data that has to drive the vehicles. Vaarst is committed to unlocking the potential of offshore robotics for all.

He continues: "We're tremendously excited about the future, and really delivering our customers' digital and robotic ambitions."

Vaarst will operate globally, with headquarters in Bristol, and has 29 employees with plans to grow to 70+ by end of 2022. The company is a technology spin-off of Rovco which was founded in 2015 and has invested heavily in real-time artificial intelligence-based 3D vision and autonomy systems. Future plans will see Vaarst take its offering to the wider industrial robotics global markets in sectors such as mining, construction, farming and land survey.

Here is the original post:

Vaarst launches to drive the future of marine robotics through data focus | RoboticsTomorrow - Robotics Tomorrow

Abstract

Artificial microswimmers that can replicate the complex behavior of active matter are often designed to mimic the self-propulsion of microscopic living organisms. However, compared with their living counterparts, artificial microswimmers have a limited ability to adapt to environmental signals or to retain a physical memory to yield optimized emergent behavior. Different from macroscopic living systems and robots, both microscopic living organisms and artificial microswimmers are subject to Brownian motion, which randomizes their position and propulsion direction. Here, we combine real-world artificial active particles with machine learning algorithms to explore their adaptive behavior in a noisy environment with reinforcement learning. We use a real-time control of self-thermophoretic active particles to demonstrate the solution of a simple standard navigation problem under the inevitable influence of Brownian motion at these length scales. We show that, with external control, collective learning is possible. Concerning the learning under noise, we find that noise decreases the learning speed, modifies the optimal behavior, and also increases the strength of the decisions made. As a consequence of time delay in the feedback loop controlling the particles, an optimum velocity, reminiscent of optimal run-and-tumble times of bacteria, is found for the system, which is conjectured to be a universal property of systems exhibiting delayed response in a noisy environment.

Living organisms adapt their behavior according to their environment to achieve a particular goal. Information about the state of the environment is sensed, processed, and encoded in biochemical processes in the organism to provide appropriate actions or properties. These learning or adaptive processes occur within the lifetime of a generation, over multiple generations, or over evolutionarily relevant time scales. They lead to specific behaviors of individuals and collectives. Swarms of fish or flocks of birds have developed collective strategies adapted to the existence of predators (1), and collective hunting may represent a more efficient foraging tactic (2). Birds learn how to use convective air flows (3). Sperm have evolved complex swimming patterns to explore chemical gradients in chemotaxis (4), and bacteria express specific shapes to follow gravity (5).

Inspired by these optimization processes, learning strategies that reduce the complexity of the physical and chemical processes in living matter to a mathematical procedure have been developed (6). Many of these learning strategies have been implemented into robotic systems (79). One particular framework is reinforcement learning (RL), in which an agent gains experience by interacting with its environment (10). The value of this experience relates to rewards (or penalties) connected to the states that the agent can occupy. The learning process then maximizes the cumulative reward for a chain of actions to obtain the so-called policy. This policy advises the agent which action to take. Recent computational studies, for example, reveal that RL can provide optimal strategies for the navigation of active particles through flows (1113), the swarming of robots (1416), the soaring of birds (3), or the development of collective motion (17). The ability of how fish can harness the vortices in the flow field of others for energy-efficient swimming has been explored (18). Strategies of how to optimally steer active particles in a potential energy landscape (19) have been explored in simulations, and deep Q-learning approaches have been suggested to navigate colloidal robots in an unknown environment (20).

Artificial microswimmers are a class of active materials that integrate the fundamental functionality of persistent directed motion, common to their biological counterparts, into a user-designed microscopic object (21). Their motility has already revealed insights into a number of fundamental processes, including collective phenomena (2224), and they are explored for drug delivery (25) and environmental purposes (26). However, the integration of energy supply, sensing, signal processing, memory, and propulsion into a micrometer-sized artificial swimmer remains a technological challenge (27). Hence, external control strategies have been applied to introduce sensing and signal processing, yet only schemes with rigid rules simulating specific behaviors have been developed (2831). Combining elements of machine learning and real-world artificial microswimmers would considerably extend the current computational studies into real-world applications for the future development of smart artificial microswimmers (32).

Here, we incorporate algorithms of RL with external control strategies into the motion of artificial microswimmers in an aqueous solution. While the learning algorithm is running on a computer, we control a real agent acting in a real world subjected to thermal fluctuations, hydrodynamic and steric interactions, and many other influences. In this way, it is possible to include real-world objects in a simulation, which will help to close the so-called reality gap, i.e., the difference of pure in silico learning and real-world machine learning even at microscopic length scales (27). Our experimental investigation thus goes beyond previous purely computational studies (3, 1113, 20). It allows us to observe the whole learning process optimizing parameters, which are not accessible in studies of biological species, to identify the most important ingredients of the real dynamics and to set up more realistic, but still simple, models based on this information. It also provides a glimpse of the challenges of RL for objects at those length scales for future developments.

To couple machine learning with microswimmers, we used a light-controlled self-thermophoretic microswimmer with surface-attached gold nanoparticles (Fig. 1A and see the Supplementary Materials). For self-propulsion, the swimmer has to break the time symmetry of low Reynolds number hydrodynamics (33). This is achieved by an asymmetric illumination of the particle with laser light of 532-nm wavelength. It is absorbed by the gold nanoparticles and generates a temperature gradient along their surface, inducing thermo-osmotic surface flows and lastly resulting in a self-propulsion of the microswimmer suspended in water. The direction of propulsion is set by the vector pointing from the laser position to the center of the particle. The asymmetric illumination is maintained during the particle motion by following the swimmers position in real time and steering the heating laser (see the Methods section below). As compared with other types of swimmers (28, 34, 35), this symmetric swimmer removes the time scale of rotational diffusion from the swimmers motion and provides an enhanced steering accuracy (36, 37) (see the Supplementary Materials).

(A) Sketch of the self-thermophoretic symmetric microswimmer. The particles used have an average radius of r = 1.09 m and were covered on 30% of their surface with gold nanoparticles of about 10 nm diameter. A heating laser illuminates the colloid asymmetrically (at a distance d from the center), and the swimmer acquires a well-defined thermophoretic velocity v. (B) The gridworld contains 25 inner states (blue) with one goal at the top right corner (green). A set of 24 boundary states (red) is defined for the study of the noise influence. (C) In each of the states, we consider eight possible actions in which the particle is thermophoretically propelled along the indicated directions by positioning the laser focus accordingly. (D) The RL loop starts with measuring the position of the active particle and determining the state. For this state, a specific action is determined with the greedy procedure (see the Supplementary Materials for details). Afterward, a transition is made, the new state is determined, and a reward for the transition is given. On the basis of this reward, the Q-matrix is updated, and the procedure starts from step 1 until an episode ends by reaching the goal or exiting the gridworld to a boundary state.

To show RL with a real-world microscopic agent, we refer to the standard problem of RL, the gridworld. The gridworld problem allows us to have an experimental demonstration while being able to access the problem numerically. We coarse grain a sample region of 30 m by 30 m into a gridworld of 25 states (s, 5 5), each state having a dimension of 6 m by 6 m (Fig. 1B). One of the states is defined as the target state (goal), which the swimmer is learning to reach. The gridworld is surrounded by 24 boundary states according to Fig. 1B. The obtained real-time swimmer position is used to identify the state s in which the swimmer currently resides. To move between states, we define eight actions a. The actions are carried out by placing the heating laser at the corresponding position on the circumference of the particle (see Fig. 1C). A sequence of actions defines an episode in the gridworld, which ends when the swimmer either leaves the gridworld to a boundary state or reaches the target state. During an episode, rewards or penalties are given. Specifically, the microswimmer gets a reward once it reaches the target state and a penalty in other cases (see the Supplementary Materials for details on the reward definitions). The reward function R thus only depends on the state s, i.e., R = R(s).

We have implemented the model-free Q-learning algorithm to find the optimal policy that solves the navigation problem (38). The gained experience of the agent is stored in the Q-matrix (10), which tracks the utilities of the different actions a in each state s. When the swimmer transitions between two states s and s (see the Supplementary Materials for details on the choice of the next state), the Q-matrix is updated according toQt+t(s,a)=Qt(s,a)+[R(s)+maxaQt(s,a)Qt(s,a)](1)taking into account the reward R(s) of the next state, the utility of the next state Qt(s, a) after taking the best action a, and the current utility Qt(s, a). The influence of these values is controlled by two factors, the learning rate and the discount factor . The learning rate defines the fraction at which new information is incorporated into the Q-matrix, and the discount factor determines the value of future events into the learning process. The reward function is the only feedback signal that the system receives to figure out what it should learn. The result of this RL procedure is the optimal policy function *(s) a, which represents the learned knowledge of the system, *(s) = argmaxaQ(s, a), Q(s,a)=limtQt(s,a). Figure 1D highlights the experimental procedure of actuating the swimmer and updating the Q-matrix. As compared with computer models solving the gridworld with deterministic agents, there are four important differences to note. (i) The swimmer can occupy all positions within each state of 6 m by 6 m size. It can be arbitrarily close to the boundary. (ii) The swimmer moves in several steps through each state before making a transition. A swimmer velocity of v = 3 m s1 leads to a displacement of about 6 m within 2 s, corresponding to about 11 frames at an inverse frame rate texp = 180 ms until a transition to the next state is made. (iii) The new state after a transition does not have to be the state that was targeted by the actions. The microswimmers are subject to Brownian motion with a measured diffusion coefficient of D = 0.1 m2 s1. The trajectory is therefore partially nondeterministic. With this respect, the system we consider captures a very important feature of active matter on small length scales that is inherent to all microscopic biological systems, where active processes have been optimized to yield robust functions in a noisy background. (iv) Due to a time delay in the feedback loop controlling the active particles, the action applied to the swimmer is not determined from its present position but from its position in the past, which is a common feature for all living and nonliving responsive systems.

Figure 2 summarizes the learning process of our microswimmer for boundary states with R = 0 and a velocity of v = 3.0 m s1, v = r e /texp where r e is the mean projected displacement of the swimmer along the direction of the action e. Over the course of more than 5000 transitions (more than 400 episodes, about 7 hours of experiment), the sum of all Q-matrix entries converges (Fig. 2A). During this time, the mean number of transitions to reach the goal state decreases from about 600 transitions to less than 100 transitions (Fig. 2B). Accordingly, the trajectories of the swimmer become more deterministic, and the swimmer reaches the goal state independent of the initial state (Fig. 2C and inset). As a result of the learning process, the initial random policy is changing into a policy driving the swimmer toward the goal state. In this respect, the final policy provides an effective drift field with an absorbing boundary at the goal state (Fig. 2D). During this process, which correlates the actions of neighboring cells, the average projected velocity v causing the drift toward the goal also increases. Although the obtained policy is reflecting the best actions only, the Q-matrix shown in Fig. 2E provides the cumulative information that the swimmer obtained on the environment. It delivers, for example, also information on how much better the best action in a state has been as compared with the other possible actions. The representation in Fig. 2E encodes the Q-matrix value in the brightness of eight squares at the boundary of each state (center square has no meaning). Brighter colors thereby denote larger Q-matrix value.

(A) Learning progress for a single microswimmer in a gridworld at a velocity of v = 3.0 m s1. The progress is quantified by the sum of all Q-matrix elements at each transition of the learning process. The Q-matrix was initialized randomly. The shaded regions denote a set of 25 episodes in the learning process, where the starting point is randomly chosen. (B) Mean number of steps required to reach the target when starting at the lower left corner as the number of the learning episodes increases. (C) Different examples of the behavior of a single microswimmer at different stages of the learning process. The first example corresponds to a swimmer starting at the beginning of the learning process at an arbitrary position in the gridworld. The trajectory is characterized by a large number of loops. With an increasing number of learning episodes, the trajectories become more persistent in their motion toward the goal. This is also reflected by the decreasing average number of steps taken to reach the goal [see (B)]. The inset in the rightmost graph reveals trajectories from different starting positions. (D) Policies (s) = argmaxaQt(s, a) defined by the Q-matrix before (Qt(s, a) = Q0(s, a)) and after (Qt(s, a) = Q(s, a)) the convergence of the learning process. (E) Color representation of the initial and the final Q-matrix for the learning process. The small squares in each state represent the utility of the corresponding action (same order as in Fig. 1C) given by its Q-matrix entry, except for the central square. Darker colors show smaller utility, and brighter colors show a better utility of the corresponding action.

Because our gridworld is overlayed to the real-world sample, we may also define arbitrary obstacles by providing penalties in certain regions. Figure 3 (A and B) shows examples for trajectories and policies where the particles have been trained to reach a goal state close to a virtual obstacle. Similarly, real-world obstacles can be inserted into the sample to prevent the particle from accessing specific regions and thus realizing certain actions. More complex applications can involve the emergence of collective behavior, where the motion of multiple agents is controlled simultaneously (30). Different levels of collective and cooperative learning may be addressed (14, 39). A true collective learning is carried out when the swimmer is taking an action to maximize the reward of the collective, not only its individual one. Swimmers may also learn to act as a collective when positive rewards are given if an agent behaves like others in an ensemble (17). This mimics the process of developing swarming behavior implicated, for example, by the Vicsek model (40). Our control mechanism is capable of addressing multiple swimmers separately such that they may also cooperatively explore the environment. Instead of a true collective strategy, we are considering a low density of swimmers (number of swimmers number of states), which share the information gathered during the learning process by drawing their actions from and updating the same Q-matrix. The swimmers are exploring the same gridworld in different spatial regions, and thus, a speedup of the learning is expected. Figure 3C displays the trajectories of two particles sharing the same Q-matrix, which is updated in each learning step. As a result, the learning speed is enhanced (Fig. 3D). The proposed particle control therefore provides the possibility to explore a collective learning or the optimization of collective behavior and thus delivers an ideal model system with real physical interactions.

(A) Example trajectories for a learning process with a virtual obstacle (red square, R = 100) next to the goal state (R = 5) in the center of the gridworld. (B) Example trajectory for an active particle that has learned to reach a goal state (R = 5) behind a large virtual obstacle (red rectangle, R = 100). (C) Example trajectories for two particles sharing information during the learning process. The same rewards as in Fig. 2 have been used. (D) Sum of all Q-matrix elements at each transition comparing the learning speed with two particles sharing the information. In all the panels, the active particle speed during the learning process has been v = 3.0 m s1.

A notable difference between macroscopic agents, like robots, and microscopic active particles is the Brownian motion of microswimmers. There is an intrinsic positional noise present in the case of active particles, which is also of relevance for small living organisms like bacteria, cells, and all active processes on microscopic length scales. The advantage of the presented model system, however, is that the influence of the strength of the noise can be explored for the adaption process and the final behavior, whereas this is difficult to achieve in biological systems.

The importance of the noise in Brownian systems is commonly measured by the Peclet number, Pe = rv/2D, comparing the product of particle radius r and the deterministic particle displacement vt to the corresponding square displacements by Brownian motion 2Dt. To explore the influence of the noise strength, we change the speed of the active particle v, whereas the strength of the noise is given by the constant diffusion coefficient D. We further introduce a penalty in the boundary states R = 100 to modify the environment in a way that the influence of noise can introduce quantitative consequences for the transitions.

When varying the speed v between 2 and 5 m s1, we make four general observations. (i) Due to time delay in the feedback loop controlling the particles, the noise influence depends on the particle speed nonmonotonously (Fig. 4E and the Supplementary Materials). As a result, we find an optimal particle speed for which the noise is least important, as discussed in more detail in the following section. For the parameters used in the experiment, the optimal velocity is close to the maximum speed available. When increasing the speed in the limited interval of the experiment, the importance of the noise thus decreases. (ii) The Q-matrix converges considerably faster for higher particle speeds corresponding to a lower relative strength of the noise. This effect is intuitive because the stronger the noise, the lower the correlation between action and desired outcome. Figure 4A shows the convergence of the sum of the Q-matrix elements (summed over all entries for a given transition) for different microswimmer speeds (v = 2.8 m s1, v = 4.0 m s1, and v = 5.1 m s1). Although the sum reaches 50% after 250 transitions for the highest velocity, this requires almost 10 times more transitions at about half the speed. (iii) The resulting optimal policy depends on the noise strength. In Fig. 4B, we show the policies obtained for two different velocities (v = 1.6 m s1 and v = 4.6 m s1). Differences in the two policies are, in particular, visible in the states close to the boundary. Most of the actions at the top and right edge of the low-velocity policy point inward, whereas actions parallel to the edge are preferred at the higher velocity (see highlighted regions in Fig. 4, B and C). (iv) The contrast between the best action and the average of the other actions, which we take as a measure of the decision strength, is enhanced upon increasing importance of the noise. This contrast for a given state sk is measured byG(sk)=1{Q(sk,ab)Q(sk,ai)i}(2)where ab denotes the best action for the state and Q(sk,ai)i=i=18Q(sk,ai)/8. The result is normalized by a factor to make the largest contrast encoded in the color of the states in Fig. 4B equal to one.

(A) Sum of the Q-matrix elements as a function of the total number of transitions during the learning process. The different curves were obtained for learning with three different microswimmer speeds. (B) Policy obtained from learning processes at high noise (low velocity) (1 : v = 1.6 m s1) and low noise (high velocity) (2 : v = 4.6 m s1). The coloring of the states corresponds to the contrast between the value of the best action and the average of all other actions (Eq. 2). (C) Transition probabilities used in Bellmans Eq. 3 for diagonal and nondiagonal actions as determined from experiments with 500 trajectories for a velocity of 1.6 and 4.6 m s1. The blue lines indicate example experimental trajectories, which yield equivalent results for actions a2, a4, a5, a7 (top) and a1, a3, a6, a8 (bottom). The blue dots mark the first point outside the grid cell. The histograms to the right show the percentage arriving in the corresponding neighboring states. The numbers below denote the percentages for the two velocities (value in parentheses for higher velocity). (D) Origin of directional uncertainty. The green dots indicate the possible laser position due to the Brownian motion of the particle within the delay time t. The two graphs to the right display the experimental particle displacements of a single microswimmer within the delay time t = texp = 180 ms, when starting at the origin for two different particle velocities. (E) Variances of the point clouds in (D) parallel and perpendicular to the intended direction of motion. The dashed lines correspond to the theoretical prediction according to Eq. 4 for the perpendicular motion (2) and 2=2Dt+(cosh(2)1)v2t2 for the tangential motion with 20.23rad2, D = 0.1 m2 s1, and t = t = 180 ms. (F) Survival fraction of particles moving in the upper states at the boundary toward the goal state in policy 2 indicated in the inset. The survival has been determined from simulations for the same parameters as in (E).

Because the environment (gridworld with its rewards) stays constant for all learning processes at different velocities, all our above observations for varying particle speed are related to the importance of the noise strength. According to Bellmans equation (10)Q(s,a)=sP(ss,a)[R(s)+maxaQ(s,a)](3)

the influence of the noise on the learning process is encoded in the transition probabilities P(ss, a), i.e., the probabilities that an action a in the state s leads to a transition to the state s. This equation couples the element Q(s, a) of the optimized Q-matrix, corresponding to a state s and action a, with the discounted elements *(s)=maxaQ(s,a) of the optimal policy in the future states s and the corresponding future rewards R(s), weighted by transition probabilities P(sa, s). Using this equation, one can obtain the Q-matrix and the optimal policy by a Q-matrix value iteration procedure if the transition probabilities are known. The transition probabilities thus contain the physics of the motion of the active particle, including the noise, and decide how different penalties or rewards of the neighboring states influence the value of Q.

We have measured the transition function for the two types of transitions (diagonal and nondiagonal) using 500 trajectories in a single grid cell. To obtain the transition function, we set the starting position of all the trajectories to the center of the grid cell, carried out the specific action, and determined the state in which the particle trajectory ended up. The results are shown in Fig. 4C with exemplary trajectories and a histogram to the right. The numbers below the histograms show the corresponding transition probabilities to the neighboring state in percent for a velocity of v = 1.6 m s1 (v = 4.6 m s1 for the values in parentheses). The two velocities show only weak changes in the transition probabilities for the nondiagonal actions, which appear to be responsible for the changes in the policies in Fig. 4B. Carrying out a Q-matrix value iteration confirms the changes in the policy in the marked regions for the measured transition probability range (see the Supplementary Materials).

The advantage of our experimental system is that we can explore the detailed physical behavior of each microswimmer in dedicated experiments. To this end, we find two distinct influences of the Brownian motion as the only noise source on the microswimmers motion. Figure 4D shows the distribution of microswimmer displacement vectors within a time texp = 180 ms for two different velocities. Each displacement starts at the origin, and the point cloud reflects the corresponding end points of the displacement vectors. With increasing velocity, the particles increase their step length in the desired horizontal direction. The mean distance corresponds to the speed of the particle, and the end points are located close to a circle. At the same time, a directional uncertainty is observed where the angular variance 2 is nearly constant for all speeds (see the Supplementary Materials for details). This directional noise is the result of a delayed action in the experiments (30, 41), i.e., a time separation between sensing (imaging the position of the particle) and action on the particle position (placing the laser for propulsion). Both are separated by a delay time t, which is the intrinsic delay of the feedback loop (t = texp = 180 ms in our experiments). A delayed response is a very generic feature of all active responsive systems, including biological species. In the present case of a constant propulsion speed, it leads to an anisotropic noise. In the direction perpendicular to the intended action, the Brownian noise gets an additional component that is increasing nonlinearly with the particle speed, whereas the noise along the intended direction of motion is almost constant (Fig. 4E).

The increase in the variance perpendicular to the direction of motion can be analyzed with a simple model (see the Supplementary Materials for details), which yields2=v2tsinh(2)t+2Dt(4)and corresponds well with experimental data (Fig. 4E) for 20.23rad2 and fixed time t = t. In particular, it captures the nonlinear increase of 2 with the particle speed v.

The increase has important consequences. When considering the motion in the top four states of policy 2 (Fig. 4B), the particle would move horizontally toward the goal starting at an arbitrary position in the leftmost state. From all trajectories that started, only a fraction will arrive at the goal state before leaving these states through the upper, lower, or left boundaries of those four states. This survival fraction has been determined from simulations (also see the Supplementary Materials for an approximate theoretical description). Overall, a change between the two policies 1 and 2 is induced by an increase of the survival by less than 10% when going from v = 1.6 m s1 to v = 4.6 m s1. When further increasing the velocity, we find in simulations that an optimal velocity for maximum survival exists. This maximum corresponds to the minimumvopt=2Dsinh(2)t(5)in the variance (Eq. 4) for a fixed traveled distance a = vt, which only depends on the diffusion coefficient D, the angular variance 2, and the sensorial delay t (see the Supplementary Materials for details). In the limit of instantaneous actions (t = 0), an infinitely fast motion would yield the best results. Any nonzero delay will introduce a speed limit at which a maximum survival is ensured. We expect that the optimal policy for very high velocities should yield a similar policy as for low velocities. An experimental verification of this conjecture is currently out of reach, as Fig. 4F shows the results of the simulations.

The observed behavior of the survival probability, which exhibits a maximum for a certain particle velocity, implies that the probability to reach the target is maximal for the same optimal velocity. Moreover, because the underlying analysis is solely based on the competition of two noises omnipresent in (Brownian) active matter, namely the diffusion and the uncertainty in choosing the right direction, we conjecture that the observed type of behavior is universal. The precision of reaching the target (long time variance of the distance from the target) by the run-and-tumble motion of bacteria exhibits a minimum as a function of the run-and-tumble times (42, 43) reminiscent of our results. These results also demonstrate that the combination of machine learning algorithms with real-world microscopic agents can help to uncover physical phenomena (such as time delay in the present work), which play important roles in the microscopic motion of biological species.

Concluding, we have demonstrated RL with a self-thermophoretic microswimmer carrying out actions in a real-world environment with its information processing and sensing capabilities externalized to a computer and a microscopy setup. Already with this hybrid solution, one obtains a model system, where strategies in a noisy environment with virtual obstacles or collective learning can be explored. Although our simple realization of a gridworld is based on a global position detection defining the state of the swimmer, future applications will consider local information, e.g., the response to a temporal sequence of local physical or chemical signals, to allow for navigation in unknown environments. As compared with a computer simulation, our system contains a nonideal control limited by the finite reaction time of the feedback loop, presence of liquid flows, imperfections of the swimmers or sample container, hydrodynamic interactions, or other uncontrolled parameters that naturally influence the learning process. In this way, it resembles a new form of computer simulation using real-world agents. An important advantage is that the physics of the agent can be explored experimentally in detail to understand the learned strategies, and the real-world interactions in more complex environments can be used to adapt the microswimmers behavior. In that sense, even the inverse problem of using the learned strategy to reveal the details of these uncontrolled influences may be addressed as a new form of environmental sensing. Similarly, the control of active particles by machine learning algorithms may be used in evolutionary robotics (8, 44), where the interaction of multiple particles may be optimized to yield higher-order functional structures based on environmental interactions. Although the implementation of signaling and feedback by physical or chemical processes into a single artificial microswimmer is still a distant goal, the current hybrid solution opens a whole branch of new possibilities for understanding adaptive behavior of single microswimmers in noisy environments and the emergence of collective behavior of large ensembles of active systems.

Samples consisted of commercially available gold nanoparticlecoated melamine resin particles of a diameter of 2.19 m (microParticles GmbH, Berlin, Germany). The gold nanoparticles were covering about 30% of the surface area and were between 8 and 30 nm in diameter (see the Supplementary Materials for details.) Microscopy glass cover slides were dipped into a 5% Pluronic F127 solution, rinsed with deionized water, and dried with nitrogen. The Pluronic F127 coating prevented sticking of the particles to the glass cover slides. Two microliters of particle suspension was placed on the cover slides to spread about an area of 1 cm by 1 cm, forming a 3-m-thin water film. The edges of the sample were sealed with silicone oil (polydimethylsiloxane) to prevent water evaporation.

Samples were investigated in a custom-built inverted dark-field microscopy setup based on an Olympus IX-71 microscopy stand. The sample was held by a Piezo stage (Physik Instrumente) that was mounted on a custom-built stepper stage for coarse control. The sample was illuminated by a halogen lamp (Olympus) using a dark-field oil-immersion condenser [Olympus, numerical aperture (NA), 1.2]. The scattered light was collected by an oil-immersion objective lens (Olympus, 100, NA 1.35 to 0.6) with the NA set to 0.6 and captured with an Andor iXon emCCD camera. A = 532 nm laser was focused by the imaging objective into the sample plane to serve as a heating laser for the swimmers. Its position in the sample plane was steered by an acousto-optic deflector (AOD; AA Opto-Electronic) together with a 4-f system (two f = 20 cm lenses). The AOD was controlled by an ADwin realtime board (ADwin-Gold, Jger Messtechnik) exchanging data with a custom LabVIEW program. A region of interest of 512 pixels by 512 pixels (30 m by 30 m) was used for the real-time imaging, analysis, and recording of the particles, with an exposure time of texp = 180 ms. The details of integrating the RL procedure are contained in the Supplementary Materials.

robotics.sciencemag.org/cgi/content/full/6/52/eabd9285/DC1

Fig. S1. Symmetric swimmer structure.

Fig. S2. Swimmer speed as a function of laser power.

Fig. S3. Directional noise as function of the swimming velocity measured in the experiment.

Fig. S4. Directional noise model.

Fig. S5. Results of the analytical model of the influence of the noise.

Fig. S6. Q-matrix value iteration result.

Movie S1. Single-swimmer free navigation toward a target during learning.

Movie S2. Single-swimmer free navigation toward a target after learning.

Movie S3. Navigation toward a target with virtual obstacles.

Movie S4. Multiple-swimmer free navigation toward a target.

J. K. Parrish, W. M. Hamner, Animal Groups in Three Dimensions (Cambridge Univ. Press, 1997).

J. Kober, J. Peters, Reinforcement learning in robotics: A survey, in Learning Motor Skills (Springer Tracts in Advanced Robotics, 2014), vol. 97, pp. 967.

M. Wiering, M. v. Otterlo, Reinforcement Learning, in Adaptation, Learning, and Optimization (Springer Berlin Heidelberg, 2012), vol. 12.

R. S. Sutton, A. G. Barto, Reinforcement Learning: An Introduction (MIT Press, 1998).

J. C. H. Watkins, thesis, Kings College, Cambridge (1989).

L. Busoniu, R. Babuka, B. De Schutter, Multi-agent reinforcement learning: A survey, in Proceedings of the 9th International Conference on Control, Automation, Robotics and Vision (ICARCV 2006) (Singapore, 2006), pp. 527532.

Acknowledgments: Helpful discussion with P. Romanczuk is acknowledged in pointing out observations of directional noise for biological systems. Fruitful discussion and help with extrapolating the theory to the experiments by K. Ghazi-Zahedi are acknowledged. We thank A. Kramer for helping to revise the manuscript. Funding: The authors acknowledge financial support by the DFG Priority Program 1726 Microswimmers through project 237143019. F.C. is supported by the DFG grant 432421051. V.H. is supported by a Humboldt grant of the Alexander von Humboldt Foundation and by the Czech Science Foundation (project no. 20-02955J). Author contributions: F.C. conceived the research. S.M.-L. and F.C. designed the experiments. S.M.-L. implemented the system, and S.M.-L. and A.F. performed the experiments. S.M.-L., V.H., and F.C. analyzed and discussed the data. F.C., V.H., and S.M.-L. wrote the manuscript. Competing interests: The authors declare that they have no competing financial interests. Data and materials availability: All data needed to evaluate the conclusions are available in the paper or in the Supplementary Materials. Additional data and materials are available upon request.

Read more here:

Reinforcement learning with artificial microswimmers - Science

Generations of laboratory mice like these recently became host to microscopic robot swarms.Photo: Getty Images (Getty Images)

In a mind-bending development, a team of researchers in China have managed to treat brain tumors in mice by delivering drugs to the tissues using microscopic robots. The robots jumped from the mices bloodstreams into their brains by being coated in E. coli, which tricked the rodents immune systems into attacking them, absorbing the robots and the cancer-fighting drugs in the process.

The teams research was published today in the journal Science Robotics. It comes on the heels of previous research by members of the same team, which saw liquid-coated nanorobots remotely propelled through the jelly-like fluid of the eye. Besides being an obvious recipe for an episode of The Magic School Bus, the research had obvious applications for ophthalmological research and medical treatments.

Its not just the blood-brain barrier, said lead author Zhiguang Wu, a chemist at the Harbin Institute for Technology in China, in an email. Most barriers in dense tissues are difficult obstacles to overcome in moving microrobots around a body.

The crafts are magnetic, and the researchers use a rotating magnetic field to pull them around remotely. On microscaleswere talking incremental movements about 1% the width of a hairthe researchers were able to make the hybrid bio-bots wend paths like in the video game Snake. Theyre dubbed neutrobotsbecause they infiltrate the brain in the casing of neutrophils, a type of white blood cell.

The biggest challenge of the work was how to achieve a swarm intelligence of neutrobots, Wu said. Like robot swarms in the macroscale world, the micro/nanorobot swarms enable sophisticated manipulation to accomplish complex tasks.

G/O Media may get a commission

It ultimately took Wus team eight years to actualize the microscopic robot swarms capable of bridging the gap between the rodent bloodstream in the animals tail, where the bots were injected, and its brain, where gliomastumors that emerge from the brains glial cellsresided. Part of the issue is that the mices white blood cells didnt dig the flavor of the magnetic robots. To overcome that issue, Wus team coated the bots in bits of E. coli membrane, which the white blood cells easily recognize as a unwelcome invader. That made the robots much more palatable, and the white blood cells enveloped them. From inside those cells, the robots were then able to roll the cells toward the brain; a Trojan horse for the 21st century (in this case, one that benefits the residents of Troy). The neutrobots made it into the brains and were able to deliver the drug directly to the targeted tumors.

Wu said the applications of the robots are manifold, and more breakthroughs could be on the horizon. The neutrobots are not exclusively designed for the treatment of glioma, he said, explaining that theyre a platform for active delivery for the therapy of various brain diseases such as cerebral thrombosis, apoplexy, and epilepsy.

Whether its surgery or drug delivery, robots are slowly but surely making their way into our most personal of domains. Of course, theyre still just in mouse brains for now, but future applications in humans seemincreasingly likely.

The use of neutrophils in microrobot design is a fascinating strategy for overcoming biological barriers, wrote robotic engineers Junsun Hwang and Hongsoo Choi, who werent affiliated with the new work, in an accompanying article. However, bench-to-bedside translation with respect to targeted drug delivery by neutrobots or microrobots is still some way off.

Currently, experts lack the ability to see what the robots are doing clearly in real time, which would be vital for any medical use of the droids down the line. But in the rat race of robotics research, its clear that humans are pushing their inanimate swarms in the direction of progress.

Continue reading here:

Researchers Found a Way to Send Tiny Robots Into Mouse Brains - Gizmodo

Vaarst, a technology spin-off from leading subsea robotic and hydrographic survey company Rovco, was formally launched today with the goal of revolutionising the offshore robotics sector leveraging intelligent data flows for smart asset management and creating an energy-efficient and more sustainable future.

The business aims to accelerate advancement in ocean robotics, giving marine and subsea providers access to next-generation technology that will enable them to deliver AI-driven autonomous robotic work at scale.

Vaarst will target the energy and marine sectors through its innovative technologies, such as SubSLAM X2 an intelligent data collection system that delivers robotic spatial awareness and live 3D point clouds to any device in the world, without costly positioning systems, thereby saving many project days. This, combined with the companys machine learning and autonomy expertise will then provide the very best in efficient data collection and AI interpretation.

The new spinout company, Vaarst, is predicting immediate 2021 revenues over 1m rising to 20m+ rapidly in the next few years.

Vaarst CEO and Founder, Brian Allen, said, Autonomous robotics are the key to reducing the cost of offshore operations. At the same time, digitalisation of field assets is essential as the industry evolves, marrying these two concepts is needed to realise the real benefit of modern tech. Its the data that has to drive the vehicles. Vaarst is committed to unlocking the potential of offshore robotics for all.

Were tremendously excited about the future, and really delivering our customers digital and robotic ambitions, he added.

Vaarst will operate globally, with headquarters in Bristol, and has 29 employees with plans to grow to 70+ by end of 2022. The company is a technology spin-off of Rovco which was founded in 2015 and has invested heavily in real-time artificial intelligence-based 3D vision and autonomy systems. Future plans will see Vaarst take its offering to the wider industrial robotics global markets in sectors such as mining, construction, farming and land survey.

Sea News, March 25

Read the original here:

Vaarst to drive the future of marine robotics through data focus - Sea News