USV Dynamic Accurate Obstacle Avoidance Based on Improved Velocity Obstacle Method

Unmanned surface vehicle (USV) path planning is a crucial technology for achieving USV autonomous navigation. Under global planning, dynamic local obstacle avoidance has become the primary focus safe In this study, method based on enhanced velocity proposed in order to achieve replanning. Through further analysis of obstacles, geometric model set conventional was redefined. A special triangular reconstruct region. The collision time predicted by fitting previously gathered data detected obstacle’s distance, azimuth, and other relevant data. Then, it combined with risk determine when should begin end. ensure driving between points, international maritime rules (COLREGs) are incorporated accuracy avoidance. Finally, through numerical simulations various scenarios, determined that, under assumption ensuring encounter maximum change rates heading angle optimized 17.54%, 58.16%, 28.63% crossing, head-on, overtaking, respectively. results indicate optimizing angle, can avoid ship rollover caused an excessive during high-speed movement more accurate action event safety encounter.