首页|New Robotics Findings Has Been Reported by Investigators at Uni- versity College London (UCL) (Real-time Terrain Anomaly Percep- tion for Safe Robot Locomotion Using a Digital Double Framework)
New Robotics Findings Has Been Reported by Investigators at Uni- versity College London (UCL) (Real-time Terrain Anomaly Percep- tion for Safe Robot Locomotion Using a Digital Double Framework)
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
Elsevier
Researchers detail new data in Robotics. According to news reporting out of London, United Kingdom, by NewsRx editors, research stated, "Digital twinning systems are effective tools to test and develop new robotic capabilities before applying them in the real world. This work presents a real-time digital double framework that improves and facilitates robot perception of the environment." Financial support for this research came from Singapore government's Research, Innovation and Enter- prise 2025 Plan (RIE2025). Our news journalists obtained a quote from the research from University College London (UCL), "Soft or non-rigid terrains can cause locomotion failures, while visual perception alone is often insufficient to assess the physical properties of such surfaces. To tackle this problem we employ the proposed framework to estimate ground collapsibility through physical interactions while the robot is dynamically walking on challenging terrains. We extract discrepancy information between the two systems, a simulated digital double that is synchronized with real robot, both using exactly the same physical model and locomotion controller. The discrepancy in sensor measurements between the real robot and its digital double serves as a critical indicator of anomalies between expected and actual motion and is utilized as input to a learning-based model for terrain collapsibility analysis. The performance of the collapsibility estimation was evaluated in variety of real-world scenarios involving flat, inclined, elevated, and outdoor terrains."
LondonUnited KingdomEuropeEmerging TechnologiesMachine LearningRobotRoboticsUniversity College London (UCL)
NETL
NSTL
Elsevier
