Foot trajectory as a key factor for diverse gait patterns in quadruped robot locomotion

Four-legged robots are becoming increasingly pivotal in navigating challenging environments, such as construction sites and disaster zones. While substantial progress in robotic mobility has been achieved using reinforcement learning techniques, quadruped animals exhibit superior agility by employing fundamentally different strategies. Bio-inspired controllers have been developed to replicate and understand biological locomotion strategies. However, a comprehensive understanding of the influence of foot trajectories on gait patterns is still necessary. This study provides a groundbreaking perspective on the essential impact of these trajectory shapes on robotic gait patterns and overall performance. By employing the Unitree A1 robot model with a bio-inspired neural control system, our simulations demonstrate that specific trajectory shapes effectively replicate diverse and natural gait patterns, such as trotting, pacing, and galloping, thereby improving adaptability to diverse terrains. Specifically, trajectories designed for pacing exhibit superior performance on rough terrain, excelling in efficiency and adaptability over other gaits. This study highlights the significance of foot trajectory in augmenting robotic locomotion and establishes a new benchmark for developing advanced robots that operate effectively in unpredictable environments.