Welding multi-robot task allocation for BIW based on hill climbing genetic algorithm

To study the problem of welding robot task allocation for car Body-in-White (BIW) product line, built the mathematic model of the problem with the strategy of thinking about reasonable abstract and process decision in different stages. Associating strong points of hill climbing algorithm (HCA) local search and genetic algorithm (GA) global search, added directed hill climbing operator to genetic algorithm so as to speed up the search for optimal solution while produce every generation population. After primary allocation results were obtained, reconsidered process restrictions and expert knowledge to revise. A side-panel of some car BIW as model presentation, programmed with MatLab and Visual C++ to realize welding multi-robot task allocation. As result, obtain the suitable allocation results and the distribution of welding spots allocated to each robot on car BIW, the efficiency and rationality of welding robot task allocation are dramatically increased, and also realizes CAPP on welding robot task allocation. On the other hand, the results of welding robots task allocation can provide reference and dependence for welding robot path optimization and welding product line balance design.