One of the newest versions of its dexterous intelligent robot dog, Lite3, has now launched in Europe. It combines advanced mobility and an open modular structure for education, research and innovative entertainment.&nbsp;<iframe width="640" height="360" src="https://www.youtube.com/embed/NNkxlKLMoMM?&wmode=opaque&rel=0&enablejsapi=1&origin=https://www.wevolver.com" frameborder="0" allowfullscreen="" class="fr-draggable" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true"></iframe>The new Lite3 series offers four product models: a basic model designed for technology enthusiasts; and three more advanced versions intended for scientific study purposes. With new features that include DEEP Robotics&rsquo; latest proprietary limb joints, control systems and advanced algorithms, Lite3 has a more powerful, quick and durable movement capability. It also offers robust scalability that can be ramped up by installing more modules.<img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1685346634643-1E57E160-8E6D-4B25-BF27-ED02B809BE3D-11136-000001DE250C2148.png" style="width: 300px;" class="fr-fic fr-dib">Lite3 product parametersMajor design upgradeThe new version incorporates significant technological advances on Lite2. The latest robot dog has proprietary leg-joint technology that provides far greater strength and drive, with torque increased by 50%. The high-torque joint drive module offers extremely high torque density, response bandwidth and reversed transmission efficiency, enhanced rotation stability, and less power consumption.&nbsp;The payload of Lite3 is 40% greater than that of its predecessor. It can carry loads of up to 7.5 kilograms. Its endurance in terms of time and range has nearly doubled. The Lite3 can operate for up to 90 minutes and has a range of 5km.&nbsp;The Lite3 features an industrial-level real-time control system with depth optimization, a first for the product lineup, with overall computing power increased threefold. The Lite 3 can climb steps up to 15 cm in height, close to the current maximum for a quadruped robot of its size, and runs at a peak speed of 4ms-1. It can also perform horizontal jumps, high jumps, front flips and other similarly difficult manoeuvres. The algorithm upgrades make its movement more agile and responsive. <img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjg0ODIzNjcyNTA5LTMucG5nIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" style="width: 516px;" class="fr-fic fr-dib">Lite 3: Advanced mobilities of high jumps, horizontal jumps, and front flips<h3 id="isPasted">How can quadruped robots help with academic research?</h3>Quadruped robots can be beneficial for academic research. They differ from two-legged, wheeled or tracked robots because they can navigate multiple terrains and perform high-risk tasks without endangering human life. Quadruped robots can be used in various fields, such as locomotion and biomechanics studies, refining robotics algorithms and control systems, artificial intelligence and machine learning, and human-robot interaction. These robots can provide researchers with valuable insights and help advance scientific knowledge across multiple disciplines.DEEP Robotics&rsquo; robot dogs are already used by some universities, including Tsinghua, Peking and Edinburgh, and University College London. The new Lite3 is positioned to be affordable for research purposes as well as for robotics enthusiasts.<h3>Open modular structure</h3>The open modular structure and interfaces of the Lite3 make it highly scalable and flexible in its capabilities. Modules such as ones for Real-Time Kinematic (RTK), artificial intelligence, edge processing and other types of sensors can be readily added.&nbsp;To enhance its navigation across very challenging terrain, the Lite3 provides levelled&ndash;perception capability interfaces (SDK, API) combined with LiDAR (Light Detection and Ranging) and depth camera to develop auto-navigation, auto-stop, obstacle avoidance, visual positioning, 3D scanning and other perceptional abilities.&nbsp; To find out more about where you can purchase a Lite3 quadruped robot, visit our <a href="https://www.deeprobotics.cn/en/index/product1.html" rel="noopener noreferrer" target="_blank">website</a>.

Lite3: new technology for increased strength, agility, mobility, speed, navigation, and endurance

Advanced mobility and modular functions allow complex moves such as front flipping.

Quadruped robot with advanced front flipping mobility

Jueying X20 can complete unmanned detection and rescue tasks in harsh conditions

With the flexibility to deliver unmanned detection and rescue services, Jueying X20 is designed for the complex terrain of a post-earthquake landscape, the insides of vulnerable debris buildings, tunnel traffic accidents, and in toxic, hypoxia, and high-density smoke environments created by chemical pollution or disaster event.

Jueying X20 Robot Dog Hazard Detection Rescue Solution

This article was first published on&nbsp;<a href="https://news.mit.edu/2023/legged-robotic-system-playing-soccer-various-terrains-0403" id="isPasted" rel="noopener noreferrer" target="_blank">MIT News</a>. &nbsp;<hr>If you&#39;ve ever played soccer with a robot, it&#39;s a familiar feeling. Sun glistens down on your face as the smell of grass permeates the air. You look around. A four-legged robot is hustling toward you, dribbling with determination.&nbsp;While the bot doesn&rsquo;t display a Lionel Messi-like level of ability, it&#39;s an impressive in-the-wild dribbling system nonetheless. Researchers from MIT&#39;s Improbable Artificial Intelligence Lab, part of the Computer Science and Artificial Intelligence Laboratory (CSAIL), have developed a legged robotic system that can dribble a soccer ball under the same conditions as humans. The bot used a mixture of onboard sensing and computing to traverse different natural terrains such as sand, gravel, mud, and snow, and adapt to their varied impact on the ball&rsquo;s motion. Like every committed athlete, &quot;DribbleBot&quot; could get up and recover the ball after falling.&nbsp;Programming robots to play soccer has been an active research area for some time. However, the team wanted to automatically learn how to actuate the legs during dribbling, to enable the discovery of hard-to-script skills for responding to diverse terrains like snow, gravel, sand, grass, and pavement. Enter, simulation.&nbsp;A robot, ball, and terrain are inside the simulation &mdash; a digital twin of the natural world. You can load in the bot and other assets and set physics parameters, and then it handles the forward simulation of the dynamics from there. Four thousand versions of the robot are simulated in parallel in real time, enabling data collection 4,000 times faster than using just one robot. That&#39;s a lot of data.&nbsp;<iframe width="640" height="360" src="https://www.youtube.com/embed/bV4ek-zI3CU?&wmode=opaque&rel=0&enablejsapi=1&origin=https://www.wevolver.com" frameborder="0" allowfullscreen="" class="fr-draggable" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true"></iframe>Video: MIT CSAIL&nbsp;The robot starts without knowing how to dribble the ball &mdash; it just receives a reward when it does, or negative reinforcement when it messes up. So, it&#39;s essentially trying to figure out what sequence of forces it should apply with its legs. &quot;One aspect of this reinforcement learning approach is that we must design a good reward to facilitate the robot learning a successful dribbling behavior,&quot; says MIT PhD student Gabe Margolis, who co-led the work along with Yandong Ji, research assistant in the Improbable AI Lab. &quot;Once we&#39;ve designed that reward, then it&#39;s practice time for the robot: In real time, it&#39;s a couple of days, and in the simulator, hundreds of days. Over time it learns to get better and better at manipulating the soccer ball to match the desired velocity.&quot;&nbsp;The bot could also navigate unfamiliar terrains and recover from falls due to a recovery controller the team built into its system. This controller lets the robot get back up after a fall and switch back to its dribbling controller to continue pursuing the ball, helping it handle out-of-distribution disruptions and terrains.&nbsp;&quot;If you look around today, most robots are wheeled. But imagine that there&#39;s a disaster scenario, flooding, or an earthquake, and we want robots to aid humans in the search-and-rescue process. We need the machines to go over terrains that aren&#39;t flat, and wheeled robots can&#39;t traverse those landscapes,&quot; says Pulkit Agrawal, MIT professor, CSAIL principal investigator, and director of Improbable AI Lab.&quot; The whole point of studying legged robots is to go terrains outside the reach of current robotic systems,&quot; he adds. &quot;Our goal in developing algorithms for legged robots is to provide autonomy in challenging and complex terrains that are currently beyond the reach of robotic systems.&quot;&nbsp;The fascination with robot quadrupeds and soccer runs deep &mdash; Canadian professor Alan Mackworth first noted the idea in a paper entitled &quot;On Seeing Robots,&quot; presented at VI-92, 1992. Japanese researchers later organized a workshop on &quot;Grand Challenges in Artificial Intelligence,&quot; which led to discussions about using soccer to promote science and technology. The project was launched as the Robot J-League a year later, and global fervor quickly ensued. Shortly after that, &quot;RoboCup&quot; was born.&nbsp;Compared to walking alone, dribbling a soccer ball imposes more constraints on DribbleBot&#39;s motion and what terrains it can traverse. The robot must adapt its locomotion to apply forces to the ball to &nbsp;dribble. The interaction between the ball and the landscape could be different than the interaction between the robot and the landscape, such as thick grass or pavement. For example, a soccer ball will experience a drag force on grass that is not present on pavement, and an incline will apply an acceleration force, changing the ball&rsquo;s typical path. However, the bot&#39;s ability to traverse different terrains is often less affected by these differences in dynamics &mdash; as long as it doesn&#39;t slip &mdash; so the soccer test can be sensitive to variations in terrain that locomotion alone isn&#39;t.&nbsp;&quot;Past approaches simplify the dribbling problem, making a modeling assumption of flat, hard ground. The motion is also designed to be more static; the robot isn&rsquo;t trying to run and manipulate the ball simultaneously,&quot; says Ji. &quot;That&#39;s where more difficult dynamics enter the control problem. We tackled this by extending recent advances that have enabled better outdoor locomotion into this compound task which combines aspects of locomotion and dexterous manipulation together.&quot;On the hardware side, the robot has a set of sensors that let it perceive the environment, allowing it to feel where it is, &quot;understand&quot; its position, and &quot;see&quot; some of its surroundings. It has a set of actuators that lets it apply forces and move itself and objects. In between the sensors and actuators sits the computer, or &quot;brain,&quot; tasked with converting sensor data into actions, which it will apply through the motors. When the robot is running on snow, it doesn&#39;t see the snow but can feel it through its motor sensors. But soccer is a trickier feat than walking &mdash; so the team leveraged cameras on the robot&#39;s head and body for a new sensory modality of vision, in addition to the new motor skill. And then &mdash; we dribble.&nbsp;&quot;Our robot can go in the wild because it carries all its sensors, cameras, and compute on board. That required some innovations in terms of getting the whole controller to fit onto this onboard compute,&quot; says Margolis. &quot;That&#39;s one area where learning helps because we can run a lightweight neural network and train it to process noisy sensor data observed by the moving robot. This is in stark contrast with most robots today: Typically a robot arm is mounted on a fixed base and sits on a workbench with a giant computer plugged right into it. Neither the computer nor the sensors are in the robotic arm! So, the whole thing is weighty, hard to move around.&quot;There&#39;s still a long way to go in making these robots as agile as their counterparts in nature, and some terrains were challenging for DribbleBot. Currently, the controller is not trained in simulated environments that include slopes or stairs. The robot isn&#39;t perceiving the geometry of the terrain; it&#39;s only estimating its material contact properties, like friction. If there&#39;s a step up, for example, the robot will get stuck &mdash; it won&#39;t be able to lift the ball over the step, an area the team wants to explore in the future. The researchers are also excited to apply lessons learned during development of DribbleBot to other tasks that involve combined locomotion and object manipulation, quickly transporting diverse objects from place to place using the legs or arms.&ldquo;DribbleBot is an impressive demonstration of the feasibility of such a system in a complex problem space that requires dynamic whole-body control,&rdquo; says Vikash Kumar, a research scientist at Facebook AI Research who was not involved in the work. &ldquo;What&rsquo;s impressive about DribbleBot is that all sensorimotor skills are synthesized in real time on a low-cost system using onboard computational resources. While it exhibits remarkable agility and coordination, it&rsquo;s merely &#39;kick-off&rsquo; for the next era. Game-On!&rdquo;The research is supported by the DARPA Machine Common Sense Program, the MIT-IBM Watson AI Lab, the National Science Foundation Institute of Artificial Intelligence and Fundamental Interactions, the U.S. Air Force Research Laboratory, and the U.S. Air Force Artificial Intelligence Accelerator. The paper will be presented at the 2023 IEEE International Conference on Robotics and Automation (ICRA).<hr>&quot;Reprinted with permission of&nbsp;<a href="https://news.mit.edu/2023/legged-robotic-system-playing-soccer-various-terrains-0403" id="isPasted" rel="noopener noreferrer" target="_blank">MIT News</a>&quot; &nbsp;

Researchers created DribbleBot, a system for in-the-wild dribbling on diverse natural terrains including sand, gravel, mud, and snow using onboard sensing and computing. In addition to these football feats, such robots may someday aid humans in search-and-rescue missions. Photo: Mike Grimmett/MIT CSAIL

“DribbleBot” can maneuver a soccer ball on landscapes such as sand, gravel, mud, and snow, using reinforcement learning to adapt to varying ball dynamics.

A four-legged robotic system for playing soccer on various terrains

DEEP Robotics recently released its newest robotics technique, of autonomous coordinated multi-robot exploration; the experiment occurred in a 3,000 m2&nbsp;unknown environment with five Jueying X20 robot dogs. The experiment accomplished the coordinated exploration of eight targets along with its fully autonomous decision-making and planning. The purpose of the experiment is to seek out the infinite applicable potentials in the field of rescue and exploration. &nbsp;<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjY3NDY4NDc4NTMzLTE2Njc0Njg0Nzg1MzMucG5nIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" style="width: 766px;" class="fr-fic fr-dib">Multiple Jueying robot dogs exploring under unknown environments togetherDuring rescue and exploration, collecting information is often found insufficient in manned operations; operators often encounter difficult situations such as anoxia, toxic gas, landslip, etc. The quadruped robot can assist or replace operators to get into the complex and hazardous areas that wheel or tracked robots have difficulties accessing. The technique of coordinated quadruped robot exploration explored information and paths shared between robots, which enhanced exploration efficiency under unknown environments.&nbsp;<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjY3NDY4NTU0MTQ0LTE2Njc0Njg1NTQxNDQucG5nIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" style="width: 766px;" class="fr-fic fr-dib">Jueying X20 discovered and recognized the targetThe technique of autonomous exploration under an unknown environment is a frontier in the research of robotics, in which specific movement coordination, like robot dance, is unable to compare with its quality and difficulty. Jueying robots were equipped with a high-resolution omnidirectional camera, Lidar, and mesh network during the experiment. The robots build the map together by Lidar and locating algorithm; they use the omnidirectional camera for deep learning and target recognition. Each robot dog autonomously assigns its unexplored area in real-time by the cooperative mechanism. In the progress of exploration, robots autonomously planned workable paths for obstacle avoidance; eventually pulled off coordinated robot exploration and target recognition in a largely unknown area.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjY3NDY4NTc0NDMzLTE2Njc0Njg1NzQ0MzMucG5nIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" style="width: 766px;" class="fr-fic fr-dib">Jueying X20 robots share information through tracked signalsTang Wei, the manager of the Perception algorithm department, introduced: The key to an efficient, coordinated exploration is the cooperative algorithm. The robots share the simplest and most useful information, which can gradually reduce bandwidth communication cost and enhance the system&#39;s adaptability under Bandwidth-constrained Conditions. The research was issued 2022 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR2022).<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjY3NDY4NjAwODA0LTE2Njc0Njg2MDA4MDQucG5nIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" style="width: 766px;" class="fr-fic fr-dib">5 Juying X20 robot dogs share the boundary of explored area with each otherDEEP Robotics is a leading quadruped robot company in Asia; the self-developed &quot;Jueying&quot; series robots are already applied in power inspection, public rescue, discovery, and other applicable fields with its advanced locomotion control algorithm, environmental perception, and AI algorithm. Besides the coordinated multi-robot exploration, DEEP Robotics has tried corporative operation with other forms of robots from the Huzhou institute of Zhejiang University in the &quot;yellow dragon&quot; cave.

DEEP Robotics recently released its newest robotics technique, which is autonomous coordinated multi-robot exploration; the experiment is operated under a 3,000 m2 unknown environment with five Jueying X20 robot dogs.

Coordinated multi-robot dog exploration in unknown environment

It would be awesome to have robots that can help us do important things, like aid disaster recovery efforts or monitor the environment. In the case of quadrupeds, robots that walk on four legs, their mobility requires many software components to work together seamlessly. It&rsquo;s a drag when all your time is spent on developing lower-level infrastructure instead of focusing on high-level behaviors. <a href="https://engineering.cmu.edu/directory/bios/johnson-aaron.html" id="isPasted" target="_blank">Aaron Johnson</a>&rsquo;s <a href="https://www.cmu.edu/me/robomechanicslab/" rel="noopener" target="_blank">Robomechanics Lab</a> experienced these frustrations firsthand. His team often had to rely on simple models for doing research because existing software solutions were not open sourced, did not provide a modular framework, and lacked end-to-end functionality.&nbsp;<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjU5NzE4NTkzNzMxLTA3MjYtZW0tMS1xdWFkcnVwZWQucG5nIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" style="width: 766px;" class="fr-fic fr-dib">A four-legged robot using Quad-SDK software is shown jumping. Source:&nbsp;Carnegie Mellon University/Robomechanics Lab&nbsp;An example of innovation born out of necessity, Johnson, an associate professor of <a href="https://www.meche.engineering.cmu.edu/index.html" id="isPasted" rel="noopener" target="_blank">mechanical engineering</a>, and his team designed their own locomotion software stack, <a href="https://github.com/robomechanics/quad-sdk" rel="noopener" target="_blank">Quad-SDK</a>, a full-stack framework for agile quadrupedal locomotion. Many graduate students have contributed to the development of the software across a two-year period, but it was time well spent. Their design can simplify the development process for roboticists everywhere because it uses an open-source license, meaning the software can be used and modified as the user wishes. Quad-SDK comes ready to use, so researchers don&rsquo;t have to worry about implementing the necessary tools and infrastructure; instead, they can get right to work on exciting behaviors and applications.&nbsp;<iframe src="https://www.youtube.com/embed/kSXKjTxKpuA?&wmode=opaque&rel=0&enablejsapi=1&origin=https://www.wevolver.com" frameborder="0" allowfullscreen="" class="fr-draggable" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" style="width: 767px; height: 430px;"></iframe>Unlike other options, Quad-SDK is also compatible with the Robot Operating System (ROS). ROS is a middleware, somewhere between hardware and software, that lets different parts of a system talk to each other. For example, if a robot senses an obstacle in its path and needs to transfer information from its perception module to its decision-making module, ROS is what enables that communication. Imagine an app that doesn&rsquo;t use iOS or Android&mdash;it&rsquo;s much simpler when everything works together.Locomotion is a layered problem, as team member and Ph.D. student Ardalan Tajbakhsh describes it. &ldquo;In order to do anything meaningful on a robot, you need to have many components working together seamlessly.&rdquo; Quad-SDK provides a framework for robotics researchers and developers to focus their efforts on the core algorithms instead of the software tooling and infrastructure.<blockquote>In order to do anything meaningful on a robot, you need to have many components working together seamlessly. - Ardalan Tajbakhsh, Ph.D. student, Mechanical Engineering&nbsp;</blockquote>Other software packages are very good at solving one component, like motion planning, but it&rsquo;s critical to have end-to-end frameworks that provide the necessary algorithms, tools, and infrastructure for performing high-quality robotics research. Quad-SDK is full stack, meaning it contains every level of the hierarchy that affects quadruped locomotion, starting with global planning. This layer sits at the top of the stack; Tajbakhsh compares it to Google Maps, because it chooses where the robot should roughly go to reach its destination. The next layer, local planner, is what decides the specifics of the route, like where the robot should place its feet. This culminates in a third layer, the robot driver, which sends commands to the quadruped&rsquo;s joints to execute the desired motion.In the weeks since Quad-SDK <a href="https://leggedrobots.org/contributions.html" rel="noopener" target="_blank">won a Best Paper Award</a> at the Workshop on Legged Robots at the 2022 IEEE International Conference on Robotics and Automation (ICRA), people have already requested to use it, a clear indicator of how useful its solutions are and how the open-source format invites collaboration within the robotics community. Figuratively and literally, Quad-SDK can go many places.

Aaron Johnson’s Robomechanics Lab has developed an open-source software that enables more agile movement in legged robots.

Open-source software gives a leg up to robot research

Steep sections on slippery ground, high steps, scree and forest trails full of roots: the path up the 1,098-metre-high Mount Etzel at the southern end of Lake Zurich is peppered with numerous obstacles. But ANYmal, the quadrupedal robot from the Robotic Systems Lab at ETH Zurich, overcomes the 120 vertical metres effortlessly in a 31-minute hike. That’s 4&nbsp;minutes faster than the estimated duration for human hikers – and with no falls or missteps.This is made possible by a new control technology, which researchers at ETH Zurich led by robotics professor Marco Hutter recently presented in the journal <a target="_blank">Science Robotics</a>. “The robot has learned to combine visual perception of its environment with proprioception – its sense of touch – based on direct leg contact. This allows it to tackle rough terrain faster, more efficiently and, above all, more robustly,” Hutter says. In the future, ANYmal can be used anywhere that is too dangerous for humans or too impassable for other robots.<iframe src="https://www.youtube.com/embed/0e40Umfnl_w?&amp;wmode=opaque&amp;rel=0" frameborder="0" allowfullscreen="" class="fr-draggable" style="width: 767px; height: 431px;"></iframe>Video: Nicole Davidson / ETH Zurich&nbsp;<h2 id="isPasted">Perceiving the environment accurately</h2>To navigate difficult terrain, humans and animals quite automatically combine the visual perception of their environment with the proprioception of their legs and hands. This allows them to easily handle slippery or soft ground and move around with confidence, even when visibility is low. Until now, legged robots have been able to do this only to a limited extent.“The reason is that the information about the immediate environment recorded by laser sensors and cameras is often incomplete and ambiguous,” explains Takahiro Miki, a doctoral student in Hutter’s group and lead author of the study. For example, tall grass, shallow puddles or snow appear as insurmountable obstacles or are partially invisible, even though the robot could actually traverse them. In addition, the robot’s view can be obscured in the field by difficult lighting conditions, dust or fog.“That’s why robots like ANYmal have to be able to decide for themselves when to trust the visual perception of their environment and move forward briskly, and when it is better to proceed cautiously and with small steps,” Miki says. “And that’s the big challenge.”<h2>A virtual training camp</h2>Thanks to a new controller based on a neural network, the legged robot ANYmal, which was developed by ETH Zurich researchers and commercialized by the ETH spin-off <a href="https://www.anybotics.com/company/" target="_blank">ANYboticscall_made</a>, is now able to combine external and proprioceptive perception for the first time. Before the robot could put its capabilities to the test in the real world, the scientists exposed the system to numerous obstacles and sources of error in a virtual training camp. This let the network learn the ideal way for the robot to overcome obstacles, as well as when it can rely on environmental data – and when it would do better to ignore that data.“With this training, the robot is able to master the most difficult natural terrain without having seen it before,” says ETH Zurich Professor Hutter. This works even if the sensor data on the immediate environment is ambiguous or vague. ANYmal then plays it safe and relies on its proprioception. According to Hutter, this allows the robot to combine the best of both worlds: the speed and efficiency of external sensing and the safety of proprioceptive sensing.<h2>Use under extreme conditions</h2>Whether after an earthquake, after a nuclear disaster, or during a forest fire, robots like ANYmal can be used primarily wherever it is too dangerous for humans and where other robots cannot cope with the difficult terrain.In September of last year, ANYmal was able to demonstrate just how well the new control technology works at the <a href="https://mavt.ethz.ch/de/news-und-veranstaltungen/d-mavt-news/2021/09/siegreiche-teilnahme-an-robotik-wettbewerb.html" target="_blank">DARPA Subterranean Challenge</a>, the world’s best-known robotics competition. The ETH Zurich robot automatically and quickly overcame numerous obstacles and difficult terrain while autonomously exploring an underground system of narrow tunnels, caves, and urban infrastructure. This was a major part of why the ETH Zurich researchers, as part of the CERBERUS team, took first place with a prize of 2 million dollars.<hr><h2>Reference</h2>Miki T, Lee J, Hwangbo J, Wellhausen L: Learning robust perceptive locomotion for quadrupedal robots in the wild, Science Robotics 2022, doi: <a href="https://doi.org/10.1126/scirobotics.abk2822" target="_blank">10.1126/scirobotics.abk2822call_made</a>

The legged robot ANYmal on the rocky path to the summit of Mount Etzel, which stands 1,098 metres above sea level. (Photo: Takahiro Miki)

ETH researchers led by Marco Hutter developed a new control approach that enables a legged robot, called ANYmal, to move quickly and robustly over difficult terrain. Thanks to machine learning, the robot can combine its visual perception of the environment with its sense of touch for the first time.