Hybrid Velocity Obstacle‐Nonlinear Control Method for Real‐Time Collision Avoidance of Nonholonomic Mobile Robots

ABSTRACT Dynamic motion planning for nonholonomic wheeled mobile robots is essential for collision‐free navigation in environments containing both static and dynamic obstacles. This study presents a collision prediction‐based switching algorithm that integrates the velocity obstacle method in velocity change space (VCS‐VO) with a nonlinear stabilizing control strategy for dynamic collision avoidance. The proposed Hybrid Velocity Obstacle‐Nonlinear Control (HVO‐NLC) approach provides an effective solution for real‐time dynamic motion planning. Extensive simulations and real‐time experiments were conducted across various scenarios and compared against the VCS‐VO‐based motion planning methods. Across all tests, HVO‐NLC consistently outperformed VCS‐VO, with an average of 70% efficiency improvement in terms of excess time relative to the theoretical minimum time and 30% reduction in system oscillations, ensuring smoother and more effective navigation.