基于视觉伺服的水下机器人导引技术

Visual-based docking system for underwater vehicle

自主水下载具(AUV)需要通过停靠入站进行休眠、电池充电、传输数据等任务的需求。AUV的导引技术对延长AUV的工作时长、扩大作业范围有重要意义。在入站过程的最后阶段,AUV需要在容许范围内保持自身与对象物的相对位姿(包括位置和方向),进而完成对接。因此,确保传感器单元和控制系统在真实的水下环境中对各种干扰具有高度精确性和鲁棒性至关重要。针对此问题,提出了一种基于视觉伺服的AUV的导引技术。首先,该技术通过模型匹配认识法确定机器人与指定标的物(对象物)之间的相对位姿来确定自己的方位。其次,通过遗传算法(GA)以完成实时的位姿认识,抵消水流、图像采集干扰带来的视觉伺服干扰。采用此方法的水下机器人控制系统分别在与东京大学合作的模拟深海水下无光环境实验,以及真实海洋环境中完成了实验验证。实验结果证明,该系统具有对低照度、高浊度引起的图像劣化以及水流引起的控制干扰具备抗干扰性和鲁棒性。

A critical challenge for autonomous underwater vehicles(AUVs)is the docking operation for applications such as sleeping under the mother ship,recharging batteries,transferring data,and new mission downloading.The final stage of docking at a docking station requires the AUV to approach while keeping the pose(position and orientation)of the vehicle within an allowable range.The appropriate pose therefore demands a sensor unit and a control system that have high accuracy and robustness against disturbances existing in a real-world underwater environment.This paper introduces a vision-based AUV docking system,incorporating a 3D model-based matching technique and a Real-time Multi-step Genetic Algorithm(GA)for the real-time estimation of the robot’s relative pose.Experimental trials were conducted in environments such as simulated deep-sea lightless pools and the actual underwater conditions in the sea,utilizing a remotely operated vehicle(ROV)equipped with dual-eye cameras and a 3D marker.The experimental results affirm the capability of the proposed system to deliver high homing accuracy and resilience against disturbances affecting not only the captured camera images but also the vehicle’s movements.This achievement of successful docking via stereo vision,a novelty in the underwater vehicle environment,validates the efficacy of the proposed system for AUVs.