Optimal Robot Movement Using Modified Genetic Algorithm

This paper presents a modified genetic algorithm to calculate the optimal robot arm movement for processing a considerable commitment of tasks. In this scheduling problem, the objective is to minimize the total processing time related to tasks distances from each other. Traditional methods such as mathematical programming and branch and bound methods are very slow and inefficient to solve this problem. In our proposed algorithm, the crossover rate is large at first and gradually it is decreased based on convergence improvement in next generations. If the resulted best cost converges to global minima, the crossover rate will be decreased in next generation. This method is studied in terms of operation time, convergence speed and quality of the results. High promising results demonstrate that our proposed method is very efficient and can obtain higher quality solutions with better computational capability

Keywords:

Task-sequencing , Single robot arm movement , Genetic algorithm , Optimization , Reduced crossover