Radiation avoiding algorithm for nuclear robot path optimization

Abstract

Improved elephant herding-ant colony hybrid algorithm (IEHO-ACO) is proposed for the nuclear robot path planning problem under the dual constraints of distance and radiation in the radiation environment. In this algorithm, the particle swarm algorithm speed update method and “hybrid operation” are introduced to increase the ability of the elephant herding optimization to jump out of the local optimum. The shortest path obtained by the improved elephant herding optimization is converted into the increment of the initial pheromone distribution of the ant colony optimization to increase the convergence speed of the algorithm and reduce the “blindness” of the ants. At the same time, the “Z” transformation strategy is introduced to dynamically adjust the combination of pheromone heuristic factor and expectation heuristic factor, so as to avoid the occurrence of premature stagnation and falling into local optimum in the search process. Taking full account of the particularity of the radiation environment, the way heuristic factors and pheromone updates in the ant colony algorithm are adapted. Also, the attraction of the target point to the ants is increased when the fallback strategy solves the deadlock problem. The simulation results show that the paths obtained by IEHO-ACO are more reliable and converge faster, and it can be applied to complex environments, especially in environments containing concave obstacles, IEHO-ACO shows good search performance.