Legged Robots

Coordinated Movement has always been one of the primary challenges to tackle in robotics. Locomotion and manipulation are the two main concepts of movement. Locomotion describes the method a robot uses to transport itself from one place to another. This can take the form of legged, wheeled, or tracked motion. Manipulation, on the other hand, describes the manner in which a robot interacts with objects around it. That includes grabbing an object, picking something up, throwing something, or even opening a door.For years, researchers and engineers have studied ways to control locomotion and manipulation simultaneously. Particularly for legged robots, attaching an arm can enable them to perform several mobile manipulation tasks that wheeled or tracked robots cannot perform. The conventional way of controlling such legged manipulators has been to control locomotion and manipulation separately - otherwise known as decoupling mobility and manipulation. While this method has its perks, it does require a lot of engineering efforts to maintain sufficient coordination between the legs and the arm.&nbsp;&nbsp;<iframe id="638d493a4c420f000872b315" width="250" class="specs-iframe" height="400" src="https://wevolver.com/iframe/specs/unitree-robotics-go1-edu-plus?iframe=true" frameborder="0" scrolling="true" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true"></iframe>Researchers refer to modeling an arm-carrying legged robot as &ldquo;a dynamic, high-degree-of-freedom, and non-smooth control problem.&rdquo; This means the complexity of such a model would limit its operation to within specified settings. Venturing into different real-life scenarios would make it increasingly complex and burdensome, and errors can arise, leading to undesirable and unnatural movements.This has driven engineers to explore different methods that can bring synergy between the limbs in a way that resembles biologically coordinated movement. Think of the way your arms and legs move and how interconnected they are. For instance, it is difficult to move your left arm and your left leg in different motions while standing. The same applies to your right side. Or consider the activity of picking up something on the ground. Your legs and arm would work together to help you reach the object and pick it up. This is referred to as interlimb neural coupling. In simpler terms, it is called whole-body control. It allows for coordination between the limbs and the extension of the capabilities of individual parts. In a similar sense, researchers at Carnegie Mellon University have successfully developed a unified policy that governs an arm-carrying legged robot in a whole-body control manner using the concept of reinforcement learning (RL). This allowed them to train the robot to synergize its locomotion and manipulation and perform specific tasks through a dynamic and agile behavior.<iframe width="640" height="360" src="https://www.youtube.com/embed/t6NoCx_zitM?autoplay=1&mute=1&wmode=opaque&rel=0" 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><h2 dir="ltr" id="isPasted">A Unified Policy for Coordinated Locomotion and Manipulation</h2>Earlier this year, Zipeng Fu, Xuxin Cheng, and Deepak Pathak from Carnegie Mellon University[1] set out to test whether a learning-based, unified policy would outperform existing decoupling techniques in controlling a robot&rsquo;s movement and ability to interact with its surrounding. As they explained, hierarchy-based decoupled and semi-coupled control methods have been ineffective due to a &ldquo;lack of coordination between the arm and legs, error propagation across modules, and slow, non-smooth and unnatural motions.&rdquo;The unified policy was based on a machine-learning training concept called reinforcement learning (RL). This method simply rewards desirable behaviors and punishes undesirable ones. Through RL, the robot would learn by trial and error to perceive and analyze its environment and act accordingly.The study comprised simulation experiments, which were then transferred into real-world experimentation. However, three distinct issues made it challenging beyond the standard RL model of training in simulation and transferring to the real world:<ol><li dir="ltr">Multiple degrees of freedom (12 DoFs in the quadruped robot and 6 DoF in the attached arm)</li><li dir="ltr">Conflicting objectives in mobility and weight balance</li><li dir="ltr">Dependency between manipulation and locomotion&nbsp;&nbsp;</li></ol>This is why Fu et al. proposed adding&nbsp;regularized online adaptation&nbsp;to bridge the so-called Sim2Real gap - i.e., the information gap between the skills acquired in simulation and those transferred to the real-world system; also referred to as the&nbsp;realizability gap. Regularization is a machine learning technique that helps calibrate the model and minimize the deviation between the model and the dataset, thus minimizing the realizability gap. This adaptation helps eliminate the conventional, two-phase, teacher-student scheme used in previous Sim2Real transfer applications. This is where a &ldquo;teacher network&rdquo; is trained by RL in simulation using full-state information, and a &ldquo;student network&rdquo; attempts to imitate it in real life based on partial information from onboard observations.&nbsp;<iframe id="6389f2641a0dc20008fdb9d3" width="250" class="specs-iframe" height="400" src="https://wevolver.com/iframe/specs/interbotix-widowx-250-6dof?iframe=true" frameborder="0" scrolling="true" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true"></iframe>The researchers also tackled the dependency issue by proposing the concept of&nbsp;advantage mixing. Advantage functions describe how much a particular action is a good or bad decision given a specific state. Knowing that locomotion tasks mostly rely on leg activity and manipulation tasks on arm activity, mixing the advantage functions of both locomotion and manipulation can help encourage and boost the robot&rsquo;s learning of the unified policy.<h2 dir="ltr">A Low-Cost Legged-Robot with Whole-Body Control</h2>The researchers presented a simple design for the legged manipulator made of the following elements:<ol><li dir="ltr">A 4-legged robot platform</li><li dir="ltr">A robot arm on top of the platform</li><li dir="ltr">An RGB camera positioned next to the gripper of the arm</li><li dir="ltr">An onboard battery to power the arm and the quadruped robot</li></ol><h2 dir="ltr">Robust Unified Policy in Simulation and Real World</h2>With this design, they were able to show in the simulation experiments an increase in whole-body coordination as the legs helped the arm reach beyond its own workspace by bending to reach lower objects (e.g. picking up a cup) and standing up high to reach higher objects (e.g. wiping a whiteboard). Similarly, the arm helped the legs maintain balance even under relatively large disturbances (e.g. running with an initial speed of 1 m/s). This demonstrates a solid unified policy with a high survival rate.Advantage mixing also helped the policy focus on each task and then combine them to induce a mechanism that boosts training. Furthermore, the proposed regularized adaptation resulted in improved performance and better prediction of the environment, as errors at the level of the gripper dropped by 20% compared to other systems, such as the two-phase teacher-student scheme.As for the real-world experiments, the researchers studied three activities: teleoperation using joysticks, closed-loop control through RGB vision-guided tracking, and open-loop response to human demonstrations.<ul><li dir="ltr">For teleoperation, the researchers used two joysticks to command the end-effector (gripper) to reach particular points even beyond the space of training. They found out that both the leg and arm joints coordinated to help the end-effector reach those points.</li><li dir="ltr">During vision-guided tracking, the RGB camera was used alongside the joystick control to provide visual feedback to the robot and improve its picking tasks. It showed repeated success on both easy and hard tasks.</li><li dir="ltr">In the open-loop control experiments, the researchers tested the coupling of the robot&rsquo;s agile locomotion with its dynamic arm movement. They commanded the end-effector to follow a predefined path while the robot was walking. The results showed dynamic coordination even while walking on an uneven grass area.</li></ul><iframe width="640" height="360" src="https://www.youtube.com/embed/zhpI3JTqQGI?&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> <h2 dir="ltr">How the Unified Policy Approach be Further Expanded</h2>The success of the unified policy opens up a new scope of coordinated robotic movement. As the researchers suggested, those results are merely preliminary. Further exciting research is expected to stem from this work, such as enabling the robot to climb on a table using its front legs to pick something up from atop the table. Another suggestion was to mount a camera onto the center of the torso, providing different visual feedback, which would pave the road for vision-based policy learning.<a href="https://manipulation-locomotion.github.io/" target="_blank">Read the full study and learn more about the exciting research of Fu et al. by visiting their dedicated website.</a>1.&nbsp;Zipeng , F., Cheng, X. and Pathak, D. (2022) &ldquo;Deep Whole-Body Control: Learning a Unified Policy for Manipulation and Locomotion ,&rdquo;&nbsp;CoRL 2022 Conference Paper&nbsp;[Preprint].&nbsp;

Could this be the time we say goodbye to robotic awkwardness and hello to seamless motion? This article highlights the groundbreaking concept of whole-body control, a coordinated robotic motion technique that resembles natural movement. Imagine a robot smoothly navigating its environment, performing tasks with grace and precision. This innovative approach allows for improved efficiency and greater versatility in real-world settings.


Unifying Robotic Arm and Leg Movement for Whole-Body Control

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

A praying mantis for the moon, Mars and meteorites: the mantis-like robot "Mantis" is said to be able to independently explore planets and build an infrastructure with its claws. The hardware and software are developed by scientists from the Robotics Innovation Center at the German Research Center for Artificial Intelligence (DFKI) and the University of Bremen.&nbsp;Thanks to his bio-inspired body and various running patterns, he will master steep craters and boulder fields as well as manipulate with his front legs.&nbsp;By learning from his experiences, he should be able to act purposefully.&nbsp;The LIMES project that has just started is being funded with around 3.7 million euros by the German Aerospace Center (DLR).&nbsp;The term is four years.&nbsp;LIMES stands for “Learning Intelligent Motions for Kinematically Complex Legged Robots for Exploration in Space”. The name of the robot is borrowed from the mantis, which belongs to the mantis order.&nbsp;The robot morphology will be similar to that of the natural model: the system should not only be able to use the two front of its six legs to walk, but also to grab objects.&nbsp;To do this, it lifts its upper body – like the praying mantis. "The challenge is to learn running patterns for different ground conditions in which all six legs work in unison despite different tasks," explains Prof. Dr.&nbsp;Frank Kirchner, head of the DFKI Robotics Innovation Center and the Robotics working group at the University of Bremen.&nbsp;The many joints, distributed over several legs, enable many different movement patterns.&nbsp;In this way, the robot can adapt its posture to the circumstances – and maneuver safely in unfamiliar terrain. Robot selects situationally from its "behaviour library" The aim of the project is to give the robot algorithms for different walking modes.&nbsp;These can be designed for energy-saving or careful movement.&nbsp;Mantis senses stumbling blocks, craters and hills with tactile sensors that he wears in his soles.&nbsp;Using a camera in his head, he creates a 3D image of the environment.&nbsp;The data is transmitted to earth by satellite.&nbsp;The human instructs the robot which destination to head for.&nbsp;"Based on the data, Mantis selects the appropriate walking pattern from its so-called behavior library.&nbsp;It recognizes obstacles and then avoids them of its own accord,” says Sebastian Bartsch, DFKI computer scientist and project manager.&nbsp;After each action, the robot evaluates its decision.&nbsp;“For example, he asks himself: Was my energy use good?&nbsp;If not, he will act differently in a similar situation in the future,” says Bartsch. The movement patterns are pre-learned in simulations that simulate extraterrestrial conditions: gravitation, lighting conditions, gradients. Real tests follow in the DFKI's space exploration hall, a 288 square meter test site with an artificial crater landscape.&nbsp;

Inspired by the mantis: Robot "Mantis" will not only be able to walk with its front legs, but also to manipulate.

A praying mantis for the moon, Mars and meteorites: the mantis-like robot "Mantis" is said to be able to independently explore planets and build an infrastructure with its claws. 

The Praying Mantis As A Model - Six-Legged Robot To Build Infrastructure On Celestial Bodies

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.

How robots learn to hike

The Spacebit Engineering Challenge enables engineers around the world to contribute to the development of “Asagumo 2.0”, a four-legged walking rover that is set to fly to the Moon with the NASA Commercial Lunar Payload Services (CLPS) program in 2021 and 2022.Spacebit is a UK-based space startup developing space robotic technology and space data analytics tools for space exploration that includes advanced micro robotics and Artificial Intelligence (AI). Their approach to engineering is to deliver competitive cost solutions for the exploration of celestial bodies.Spacebit's goal is to create new opportunities for industry, academia, and government, advancing the future development of a new infrastructure and commercial resource exploration ‘offworld’ - &nbsp;beginning with the Moon.The company's micro-rover technologies and rover deployment systems are designed to carry scientific payloads in support of the Artemis Program space exploration and research goals. Spacebit is the only commercial company to secure two consecutive NASA Commercial Lunar Payload Services (CLPS) delivery missions to the Moon in 2021. Spacebit is one of the first commercial lunar companies that plans to fly on every NASA CLPS delivery on the market.Future versions of the rover would be sent to explore tubular cave structures – subsurface tunnels, believed to have been formed by ancient basaltic lava flows – to see if they are viable locations in which to build future lunar habitats for human presence on the moon.<iframe src="https://www.youtube.com/embed/QSGFL4W1guk?&amp;feature=emb_title&amp;ab_channel=SpacebitCompany&amp;wmode=opaque&amp;rel=0" frameborder="0" allowfullscreen="" class="fr-draggable" style="width: 788px; height: 406px;"></iframe><h2 dir="ltr">Contribute and get your name on the Moon</h2>Over the next several weeks Spacebit will invite the Wevolver community in an open format, for inputs and ideas on various engineering challenges as they develop new rover designs and technologies for space-based applications, beginning with the Moon. The technical feedback will be used to help improve and iterate Spacebit’s technology.The engineering inputs will be assessed and the most valuable contributors will be rewarded by having their name engraved on the Asagumo rover before it launches on its maiden journey to the Moon. Additionally, the winning engineer will be invited to the control room in the UK or US; or, participate in Spacebit’s remote Robotic Operations &amp; Control Center (ROCC). <div class="fr-img-space-wrap"><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjIzMzMzNTM3MDkwLVNwYWNlYml0LTZiLmpwZyIsImVkaXRzIjp7InJlc2l6ZSI6eyJ3aWR0aCI6OTUwLCJmaXQiOiJjb3ZlciJ9fX0=" style="width: 788px;" class="fr-fic fr-dib">The winner of the challenge will have their name engraved on the rover and be invited to the launch control room. Image credit: Spacebit/Wevolver.<h2 class="fr-img-space-wrap2">Asagumo</h2></div>Asagumo is a semi-autonomous micro rover built on a single-unit CubeSat frame, usually used for micro &nbsp;satellites. The &nbsp;rover weighs just 1.3 kilograms, is solar-powered and instead of wheels or tracks, is equipped with four spider-like legs. Mission 1 is a tech demo mission where Asagumo is equipped with sensors and camera technology; the rover will attempt to walk for ≤10 meters on the lunar surface, sending data back to Earth. The goal is to eventually explore the moon’s subsurface through the labyrinth of lunar lava tubes and caves. Because of its four legs rather than wheels or tracks, the rover has an advantageous design to explore parts of the Moon other rovers &nbsp;cannot reach.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjIzMjUzNTcyMTE4LVNwYWNlYml0LTEuanBnIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" style="width: 788px;" class="fr-fic fr-dib" alt="A small square gold rover robot with four black spider-like legs.">The Asagumo micro rover is built on a single-unit CubeSat frame. Image credit: Spacebit.&nbsp;<h2 class="fr-img-space-wrap2">How to contribute?</h2>Joining the discussion is simple. On Wevolver's social media channels, we will share regular updates regarding the progress of the project. Spacebit will ask for feedback on engineering challenges they have on which you can comment and engage with. To make sure you stay in the loop, follow their company profile on <a href="https://www.wevolver.com/profile/spacebit" target="_blank">Wevolver</a>.We're excited to see your feedback, and look forward to seeing the first Wevolver community member inscribed on the Moon!

Contribute to the development of a four-legged walking rover that is set to fly to the Moon with NASA in 2021 and 2022.

The Spacebit Engineering Challenge

<div data-wow-delay=".09s">A frenetic, hopping rodent is giving University of Michigan engineers new ways to think about and design two-legged robots.Biologists and engineers from U-M, Harvard University and the University of California, San Diego used information theory to, for the first time, quantify the entropy in an animal’s gait in three dimensions. The team’s findings, published in Nature Communications on Sept. 5, could help engineers make biomimetic robots that are tailored to specific environments.“This work points at a new way of building a robot that may be useful in navigating desert environments or desert planets,” said <a href="https://me.engin.umich.edu/people/faculty/ramanarayan-vasudevan" rel="noopener noreferrer" target="_blank">Ram Vasudevan</a>, U-M assistant professor of mechanical engineering and a co-author of the paper. “We want to try and build robots that are capable of thriving in all sorts of environments.”He recalled a time when NASA’s wheeled Curiosity rover got stuck in the Martian desert.“It dug a deeper and deeper hole,” Vasudevan said. “What if they had taken a look at how these rodents travel through the desert? Evolution has found a suitable solution several times over. Until now, we haven’t appreciated how useful a system it is in this context.”<iframe src="https://www.youtube.com/embed/35MSBu7-mk0?&amp;wmode=opaque" allowfullscreen="" class="fr-draggable" style="width: 800px; height: 421px;" frameborder="0"></iframe>The researchers explain their approach to studying the jerboa and its future applications to robotics. Video: Evan Dougherty, Michigan Engineering Communications &amp; Marketing.<div data-wow-delay=".09s">For biologists seeking to understand the ecological structure of animal communities, the findings also show how sudden changes in direction, gait and speed help the jerboa—a bipedal desert rodent—elude hungry predators and gain a competitive edge over their quadrupedal neighbors. The researchers compared the behavior of jerboas to jirds, which have four legs and are in the gerbil family.<h4>Entropy is the answer</h4>The jerboa is native to North Africa and Central Asia. Although bipedalism has evolved multiple times in rodents—including kangaroo rats and Australian hopping mice—jerboas are the only group to use three different gaits when moving on their long hind legs: hopping, skipping and running. They can leap more than half a meter high, which is about ten times their hip height.Talia Moore, a postdoctoral research fellow in the U-M Department of Ecology and Evolutionary Biology, was tasked with characterizing the jerboa’s locomotion and comparing it to other bipedal animals.<img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1563294792374-jerboa-gif-5.gif" style="width: 766px;" class="fr-fic fr-dib">IMAGE: &nbsp;In order to study a jerboa’s gait, the researchers had the rodent run laps around a track with a strategically placed high-speed camera and force plate. Credit: Talia Moore&nbsp;“When we put the jerboa on a force platform, every step had a different signature,” Moore said. “I thought, ‘This is so weird. It’s like they’re completely unpredictable and I’m unable to characterize it based on the techniques we have.’”She consulted with Vasudevan. “I said, ‘Ram, I wish there was a way to measure unpredictability,’” Moore recalls. “He said, ‘There is.’”They borrowed a metric of randomness from information theory—entropy. Entropy is a measure of the uncertainty of an outcome. With it, they developed a model that calculates the probability that the animal will move in a specific way.Moore and her colleagues suspected that these herky-jerky movements help jerboas foil an aerial predator’s efforts to calculate an intercept course.<h4>Jerboas vs. jirds in the wild and the lab</h4>Since they wanted to see how these animals behave in their natural habitat, they traveled to northwestern China and designed a novel set of field experiments to study whether jirds and jerboas escape predators in different ways.Combining their observations and their new model, they found that the jerboas were much more unpredictable than the jirds. Their maneuvers included momentary bursts of acceleration, kangaroo-like hopping, erratic zigzagging and acrobatic leaps.To test whether the jerboa’s unpredictability truly corresponds to increased predator-evasion success, the researchers performed behavioral experiments that measured the animals’ tendency to seek shelter, which is called thigmotaxis.Small foraging animals face a conflict between their needs to venture out for food and to stay under cover—to avoid becoming food. Due to this conflict, the amount of time an animal spends in an open area varies with its ability to evade predators. As evasiveness increases, the behavioral affinity to shelter, and so-called “open-field anxiety,” decrease.During simulated predation trials, Moore and her colleagues observed that the bipedal jerboas explored the entire experimental enclosure, while jirds tended to remain near the walled periphery. She performed a second experiment in the lab with captive-bred animals that had never seen predators before. The results from both experiments suggest that bipedalism in rodents is associated with lower thigmotaxis and greater predator evasion in open settings.<h4>Why this matters for robotics</h4>Much of the research into robot locomotion involves smooth movement in low-variability environments. But these findings demonstrate some of the limits to that perspective.For biomimetic robots to interact with the world, we’ll need to engineer unpredictability in motion.<cite>Talia Moore</cite><div data-wow-delay=".09s">“The key insight here is it’s a big engineering challenge to engineer unpredictability this has shown that it’s a worthwhile pursuit,” Vasudevan said.Added Moore, “If we want to make more biomimetic robots that interact with the world around us, we’re going to need to think about how to engineer unpredictability in motion.”The paper is titled “Unpredictability of escape trajectory explains predator 2 evasion ability and microhabitat preference of desert rodents.” Co-authors in addition to Moore and Vasudevan are Kimberly L. Cooper of the University of California, San Diego, and Andrew A. Biewener of Harvard University. The study was funded in part by a Chapman Memorial Scholarship to Moore.</div> </div></div>

IMAGE:  The jerboa - a bipedal desert rodent native to North Africa and Central Asia. Credit: Christian Baker