Robot skill adaptation improves safety in changing environments
Safety-aware Skill Adaptation for Reinforcement Learning in Dynamic Environments
Robotics
Summary
Robots learning new skills in busy or changing spaces often struggle to stay safe and stable. The authors developed a method called Dist-GPRL that helps robots adapt their movements step-by-step while avoiding obstacles and keeping smooth actions. This approach uses a smart way to guess good moves and guides the robot away from unsafe areas. They tested this on tasks with moving objects and found it worked better and safer than older methods, both in simulations and with real robots.
What this means in practice
- •For robotics developers: Improve robot manipulation skills in environments with moving objects by safely adjusting trajectories using local updates and safety-aware guidance.
- •For industrial automation teams: Enhance robotic operation safety in cluttered, dynamic factory floors by integrating distance-aware skill adaptation to reduce collisions and task failures.
Authors
A K M Nadimul Haque, Sheila Sutjipto, Marc G. Carmichael, Teresa Vidal-Calleja
Abstract
Skill adaptation frameworks based on reinforcement learning often require restrictive assumptions to maintain stability, such as fixed observations or tightly controlled exploration schedules. In cluttered and dynamic environments, however, unrestricted exploration can lead to unsafe behaviour and unstable learning, particularly when task-relevant observations lie near obstacles or involve moving objects. In this work, we present Dist-GPRL, a distance-aware and safety-guided reinforcement learning framework for structured robot skill adaptation. Building upon Gaussian Process (GP)-based skill parameterisation, our framework sequentially adapts overlapping local windows of sparse trajectory via-points rather than modifying the complete skill at every policy step. Raw policy outputs are correlated through the GP covariance structure, producing temporally coherent trajectory updates while reducing the action-space and credit-assignment difficulties associated with global trajectory adaptation. Safety is incorporated through two complementary forms of guidance. A safe-subspace prior derived from the Hausdorff Approximation Planner (HAP) biases policy exploration toward feasible regions, while dynamically updated distance field clearance and gradient rewards provide local obstacle awareness. A trajectory-kinematics similarity regulariser further preserves the demonstrated velocity and acceleration characteristics during adaptation. We evaluate the framework on two dynamic object-manipulation tasks in simulation and transfer the learned policy to real-world robot execution. Experimental results demonstrate higher task success, lower collision frequency, and more stable learning than the baselines, while preserving the kinematic characteristics of the demonstrated skill.