The Multi-objective Optimal Routing Algorithm for Vehicle Dynamic Navigation

摘要

Currently the researches on vehicle routing problem are focus on single index such as the shortest distance or shortest travel time. There are some defects to find optimal routing in this way, so it does not meet the requirement of vehicle dynamic navigation. This paper proposes a multi-objective optimal routing algorithm. It first selects one index as the fitness factor of genetic algorithm and gets several sub-optimal routings. Then it calculates every routing according to other indexes and finally finds the optimal routing. To improve the algorithm's convergence speed and accuracy, we improve the genetic algorithm, using the improved roulette algorithm to select operators and introduce a mark operator. Comparing with random algorithm and traditional genetic algorithm, the proposed algorithm can avoid falling into local optimization. It has the advantage of effective optimization, effective solution and stable performance, so it meets the requirement of vehicle dynamic navigation.