复杂动态场景下移动机器人避障算法

Real-time obstacle avoidance for mobile robots in dynamic environments with unknown or limited prior information is crucial to ensuring the safety and efficiency of autonomous navigation. However, the performance of traditional Velocity Obstacle (VO) methods heavily relies on the precise detection of obstacles by sensors, while sensor data often suffer from noise and errors, leading to reduced obstacle avoidance efficiency. To address these issues, this paper proposes a velocity obstacle algorithm integrated with dynamic trajectory prediction. Firstly, a velocity obstacle model is constructed to achieve real-time obstacle avoidance. Secondly, the Kalman filter is employed to predict obstacle trajectories, optimizing the velocity obstacle region. Finally, a velocity optimization function is designed, incorporating expected collision time, velocity smoothing constraints, directional constraints, and dynamic constraints to ensure safety, smoothness, and goal-oriented motion. Simulation results demonstrate that the proposed algorithm significantly outperforms traditional methods in terms of task completion time, path optimization, and motion efficiency, validating its effectiveness and robustness in dynamic environments.