Robot Locomotion
Overview
Robot locomotion concerns the mechanisms and control strategies that let robots walk, run, hop, roll, climb, swim, or combine these modes. Researchers use physics-based models, gait planning, model predictive control, and reinforcement learning, often combining perception or tactile sensing with proprioception to adapt to terrain and contact. Work is tested in simulation and on physical robots, with benchmarks for stability, speed, energy use, obstacle traversal, and transfer to unfamiliar settings. A central challenge is making versatile learned controllers reliable in the real world, where uncertain contacts, deformable terrain, sensor errors, and hardware limits can undermine performance.
Latest papers
233 papers31 papers in the last four weeks, down 16% on the four weeks before. 0.2% of all new papers.