Event
Proposing a Gaussian Process-Based Robot Skill Adaptation Framework
1 reports · 1 sources
A K M Nadimul Haque and other researchers proposed a robot skill adaptation framework based on Gaussian processes (GP) and sparse waypoints, which preserves the kinematic characteristics of skills by leveraging the first- and second-order analytical derivatives of trajectory poses. It includes three adaptation methods: optimization, behavioral cloning, and reinforcement learning (RL). The research was published in IEEE RA-L 2026 and presented at IROS 2026. In simulation and real-world evaluations of three tasks—opening a drawer, pushing a cube, and manipulating a horizontal bar—the authors claim that the proposed method achieved higher success rates than all baseline methods and demonstrated high cosine similarity and low velocity magnitude error.
Report timeline
- Gaussian Process-Based Robot Skill Learning and Adaptation Framework
A study published in IEEE RA-L 2026 and presented at IROS 2026 proposes a robot skill adaptation framework based on Gaussian processes (GPs) and sparse via-points, which preserves the kinematic characteristics of skills by using first- and second-order analytical derivatives of trajectory poses. The framework includes three adaptation methods: an optimization agent, a behavior cloning agent, and a reinforcement learning (RL) agent. In simulation and real-world evaluations on three tasks—opening a drawer, pushing a cube, and manipulating a horizontal bar—it outperformed all baseline methods in success rate and demonstrated high cosine similarity and low velocity magnitude error.
arXiv cs.RO ·