The following is translated from the original:
Research has driven progress in the field of robotics, and research relies heavily on effective platforms to test algorithms for robot control and navigation. Although many robotic platforms have been developed over the past few decades, most of them have shortcomings that limit their use in research environments.
Researchers at the University of California, Berkeley, recently developed the Berkeley Humanoid, a new robotic platform that can be used to train and test biomimetic robot control algorithms. The new biomimetic robot, introduced in a paper published on the preprint server arXiv, addresses and overcomes some limitations of previously introduced robot research platforms.
“After conducting many experiments on commercial robots, we have realized some of their weaknesses,” paper co-author Liao Qiyuan told Tech Xplore. “For example, some robot hardware is very expensive, while others are not specifically designed for learning-based control or research, which often means that it is ‘fragile’, easily damaged, and difficult to maintain and repair.”
The main goal of Liao and his colleagues ‘recent work is to develop a more cost-effective and scalable platform for robot research. These platforms should be low-cost, easy for researchers to produce and experiment, strong fail-resistant capabilities, easy to carry, require little maintenance, and should be manageable by a single operator.
“The Berkeley Bionic Robot Platform we developed has high performance, transparent low-level actuators and transmissions, higher reliability, and is cheap,” Liao explained. We achieve this by designing and building everything from scratch, including mechanical and electrical components.
Berkeley Humanoid is a compact, low-cost, lightweight robot that can be easily deployed in a robot laboratory. Its customized mechanical components include gearboxes and actuators, as well as various limbs and body parts. On the other hand, its electrical components are motor drives and inertial measurement unit (IMU) modules.
Researchers tested their robotic platform in a series of tests and demonstrated that it can effectively handle a variety of exercise tasks, even when controlled by basic learning-based strategies. Through this simple strategy, the robot is able to walk on different terrains, jump on one leg and two legs, and walk long distances in urban and unpaved terrains.
“We have launched a new high-performance, reliable, low-cost humanoid research platform and demonstrated amazing exercise responsibilities through simple learning-based strategies,” Liao said. “Berkeley’s bionic robots are specifically designed for academic research, rather than being deployed directly to industry. We hope to accelerate algorithm development and experimental verification.”
Preliminary tests highlight the major promise of the Berkeley Humanoid, showing that it can support dynamic walking in a variety of sports while also maintaining stability on different types of terrain. The platform will soon be used in Berkeley and other robotics research laboratories to train and test new algorithms for robot control and navigation.
“We now want to add arms to our robot so that the robot can conduct operation-related research,” Liao added. “As part of our next research, we also hope to use cameras to increase sensing capabilities to achieve better performance.”
If you want to learn more, you can click on the link below the video.
Thank you for watching this video. If you like it, please subscribe and like it. thank
Original text:https://techxplore.com/news/2024-08-humanoid-platform-robotics.html
Original YouTube:
More information: Qiayuan Liao et al., Berkeley Bionic Robots: A Learning Based Control Research Platform, arXiv (2024). DOI：10.48550/arxiv.2407.21781
Journal information: arXiv
Oil tubing: