The deep-seabed mining vehicle works in the high pressure, unknown and complex environment. How to use the mix-sensors system to describe the working environment correctly and real-time is the precondition and foundation of the vehicle's path planning. The operation performance and exploiting efficiency of the whole deepseabed mining system are decided by path planning technology. In this paper, a path planning algorithm based on a fast grid search and rolling optimization strategy is presented. The path planning simulation platform for mining vehicle is developed. Then the simulation of path planning algorithm is presented. The simulation result proves the validity of algorithm.
Deep-seabed mining vehicle is a kind of autonomous robots, this kind of robots can do a lot of heavy work under bad condition which the human beings cannot do(Peng and Yang, 2004). It is required the mining vehicle to walk on the seabed at a depth of about 6000m and collect polymetallic nodule depositing in the seabed when the mining vehicle is working. Therefore, as the dynamic center of deep-seabed mining system, deep-seabed mining vehicle is the key equipment of the whole system, and charges with the most complex and dangerous task. In allusion to the mining vehicle's own needs and the unknown but time-varying special environment, the vehicle must have the capacity to plan the point-to-point and coverage-type path, and identify the obstacles in advance, then go around them safely when appearing insurmountable obstacles. The mining vehicle works in the dynamic and time-varying environment, only relying on undersea environment information for routine and off-line global path planning but neglecting to track the change of environment, even based on the best planning, which can not guarantee the mining vehicle meet the performance indicators during actual running process.
