首页|Studies from University of Michigan Reveal New Findings on Robotics and Automati on (A Regret-informed Evolutionary Approach for Generating Adversarial Scenarios for Black-box Off-road Autonomy Systems)
Studies from University of Michigan Reveal New Findings on Robotics and Automati on (A Regret-informed Evolutionary Approach for Generating Adversarial Scenarios for Black-box Off-road Autonomy Systems)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – A new study on Robotics - Robotics and Automation is now available. According to news reporting out of Ann Arbor, Mich igan, by NewsRx editors, research stated, “Developing autonomous vehicles (AVs) that operate in diverse and demanding environments is a difficult challenge. Two fundamental tools that can accelerate this process are testing an AV in diverse simulated environments and identifying core system weaknesses.” Financial support for this research came from U.S. Army DEVCOM Ground Vehicle Sy stems Center. Our news journalists obtained a quote from the research from the University of M ichigan, “While most efforts focus on improving these tools for on-road AVs, thi s letter focuses on an analogous set of tools for off-road AVs. A method called Black-Box Adversarially Compounding Regret Through Evolution (BACRE) is proposed for identifying adversarial scenarios using an evolutionary algorithm guided by a novel regret-based metric for general navigation tasks. A black-box approach is often preferable when system complexity can be diverse, like with off-road AV s, and when whole-system testing is required. A custom simulation platform is al so provided to assist with the automated testing of AVs in diverse, unstructured environments. Numerical experiments demonstrate that BACRE’s evolutionary proce ss gradually increases scenario complexity to degrade vehicle performance (an ef fective and explainable process that comparable methods cannot achieve).”
Ann ArborMichiganUnited StatesNort h and Central AmericaRobotics and AutomationRoboticsUniversity of Michigan
NETL
NSTL
