面向无人艇自主靠泊的分层轨迹规划与试验

Layered trajectory planning and experiment for the autoberthing of unmanned surface vehicles

针对小型无人艇在自主靠泊时的轨迹规划任务,分别考虑远端和码头末端2阶段的主要影响因素,本文提出基于环境障碍、泊位及艇体动力学等多约束分析的分层人工势场轨迹规划方法,完成面向无人艇自主靠泊的轨迹规划任务。考虑小型无人艇自身体型较小、在低速靠泊状态下机动性较差、水域受限等导致的自主靠泊难题,提出基于靠泊约束分析的改进人工势场法,解决传统人工势场法中存在的弯角过大、艏向输出振荡和局部极小点问题。同时,考虑靠泊后期目标泊位及水域受限因素对无人艇艏向、速度的约束,使无人艇能够沿规划轨迹自主安全地完成靠泊任务。通过与传统人工势场法的仿真结果对比,验证该方法具有更好的轨迹规划性能;并基于“海豚-Ⅰ”号小型无人艇完成自主靠泊外场试验,检验所提出方法的有效性和可行性。

This work proposes a layered trajectory planning method based on the multi- constraint analysis of environmental obstacles and berth and small, unmanned surface vehicle (USV) dynamics. Trajectory planning in the autoberthing task of USVs is completed by considering the factors that influence remote and terminal phases. An improved APF method that accounts for the berthing constraints of USVs is proposed given that autonomous berthing is complicated by the small size and poor maneuverability under the low speed and limited water area of USVs. This method mainly resolves the over-angled corner,oscillating bow output,and local minimum point problems of traditional APF. The constraints on the speed of USVs and the direction of the bow caused by the berthing target and the limited water area in the later period of berthing are considered to enable the USV to perform autoberthing autonomously and safely along the resulting trajectory. This method is verified to have better trajectory planning performance than traditional APF methods. The autoberthing open field experiment performed with the small Dolphin-I USV proved that the proposed method is effective and feasible.