Research On Path Planning For Multi-Cranes Collaborative Lifting

With the rapid development of the lifting industry, the size and weight of hanging objects are increasing, which makes the lifting more and more difficult, however, single or two cranes cannot meet all the conditions of lifting, while multi-crane cannot be applied often because of many factors that we cannot make sure, especially the multi-mobile-crane. So far, the research on multi-crane is little, and most of the studies have focused on two mobile crawler cranes. The planning of multi-crane is very difficult due to the presence of ultra-high dimensional pose space, especially the multi-crane turning synergies. Because a lot of action contained in the input, it is difficult to obtain safe and correct lifting program. Due to the high cost of hanging objects, it cannot be imagined if a lifting accident occurs. Since the conventional methods cannot guarantee the correctness of lifting solutions, the applications of multi-crane is little, which reflects the urgency to develop new methods of lifting. Therefore, on the basis of the dual-lifting, multi-crane hoisting simulation is studied in order to get a good lifting method of providing effective reference for the actual lifting.In this paper, a mathematical model of multi-machine hoisting system is performed, and the mathematical model of multi-machine system is abstracted by using the coordinates and size constraints of expression, and analysis of multi-chain coupling constraint and the pose of multi-machine lifting systems can be obtained by the coordinate conversion formula. At the same time, the nonholonomic and kinematic constraints and performance constraints of multi-machine crane planning system are also considered.The RRTConnect++algorithm used in the dual system has been improved to make it suitable for multi-machine configuration space hoisting system. Then the turning of multi-machine system is analyzed, and an improved algorithm -- RRTConnect++variable region Algorithm is proposed. The new algorithm has been improved sample area, which greatly improves the efficiency of programming; and a new expansion strategy is also proposed in the new algorithm that the conditions of meet is enlarged, and a interpolation methods is used to check whether the path is correct. The new expansion strategy makes the path smoother and the planning more efficient.Three cases are showed, the effect of RRTConnect++programming algorithm is contrasted by new algorithm in the first case, which verifies the algorithm is effective and efficient. The second case verifies the turn of dual-crane, and four cranes are taken as an example to verify the validity of RRTConnect++variable region algorithm for multi-crane system planning in the third case.Finally, a six crane hoisting system has been modeled with a same method of four cranes and make a path planning to six crane systems with RRTConnect++variable region algorithm, which proves the RRTConnect++variable region planning algorithm is universal and efficiency for multi-crane system.