Robotics & Machine Learning Daily News2024,Issue(Jun.25) :43-43.

Recent Findings from Beijing Institute of Technology Has Provided New Informatio n about Robotics (Perception-driven Learning of High-dynamic Jumping Motions for Single-legged Robots)

北京工业大学最近的发现提供了关于机器人学的新信息(单腿机器人高动态跳跃运动的感知驱动学习)

Robotics & Machine Learning Daily News2024,Issue(Jun.25) :43-43.

Recent Findings from Beijing Institute of Technology Has Provided New Informatio n about Robotics (Perception-driven Learning of High-dynamic Jumping Motions for Single-legged Robots)

北京工业大学最近的发现提供了关于机器人学的新信息(单腿机器人高动态跳跃运动的感知驱动学习)

扫码查看

摘要

一位新闻记者兼机器人与机器学习的新闻编辑每日新闻研究人员详细介绍了机器人S的新数据。根据NewsRx记者从中华人民共和国北京发回的新闻报道,研究表明:“腿机器人在与物理环境的持续互动中表现出强大的高动态运动能力。”然而,要实现动物般的敏捷性仍然是一个重大挑战.腿类动物通常通过结合本体感觉和外部感觉的高维信息来预测和计划下一步的运动,并调整身体骨骼肌系统的硬度以适应当前的环境.本研究的资金来源包括国家重点研究计划、国家重点研究计划。我们的新闻编辑引用了北京理工学院的一篇研究文章:“传统的控制方法在处理高维状态信息或难以精确规划的复杂机器人运动方面存在局限性,深度强化学习(DRL)算法为机器人运动控制问题提供了新的解决方案。”将DRL算法与虚拟模型控制(VMC)方法相结合,提出了一种基于感知驱动的高动态跳跃自适应学习算法.该算法在了解机器人实时跳跃高度的同时,通过学习与连续跳跃有关的因素,充分挖掘机器人的运动潜能.仿真中所提出的策略在仿生单腿反渗透机器人测试平台上成功部署,无需进一步调整.

Abstract

By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Researchers detail new data in Robotic s. According to news reporting originating from Beijing, People's Republic of Ch ina, by NewsRx correspondents, research stated, "Legged robots show great potent ial for high-dynamic motions in continuous interaction with the physical environ ment, yet achieving animal-like agility remains significant challenges. Legged a nimals usually predict and plan their next locomotion by combining high-dimensio nal information from proprioception and exteroception, and adjust the stiffness of the body's skeletal muscle system to adapt to the current environment." Financial supporters for this research include The National Key Research Program of China, National Key Research Program of China. Our news editors obtained a quote from the research from the Beijing Institute o f Technology, "Traditional control methods have limitations in handling high-dim ensional state information or complex robot motion that are difficult to plan ma nually, and Deep Reinforcement Learning (DRL) algorithms provide new solutions t o robot motioncontrol problems. Inspired by biomimetics theory, we propose a per ceptiondriven high-dynamic jump adaptive learning algorithm by combining DRL al gorithms with Virtual Model Control (VMC) method. The robot will be fully traine d in simulation to explore its motion potential by learning the factors related to continuous jumping while knowing its real-time jumping height. The policy tra ined in simulation is successfully deployed on the bio-inspired single-legged ro bot testing platform without further adjustments."

Key words

Beijing/People's Republic of China/Asi a/Emerging Technologies/Machine Learning/Nano-robot/Robot/Robotics/Beijing Institute of Technology

引用本文复制引用

出版年

2024
Robotics & Machine Learning Daily News

Robotics & Machine Learning Daily News

ISSN:
段落导航相关论文