Obstacle Avoidance for Ship Navigation Safety Combining Heuristic Search Algorithm and Improved ACO Algorithm

The safety of ship navigation has always been a focus of attention in the field of maritime transport and navigation. In the complex marine environment, ships face a variety of obstacles, such as other ships, reefs, buoys, etc., which may pose a threat to navigation safety. Traditional obstacle avoidance methods mainly rely on the navigator's empirical judgement, but there are limitations and risks associated with this method. The standard ant colony optimisation algorithm tends to fall into local optimal solutions during path search, while the A* algorithm is easily limited by the search space when dealing with large-scale problems. Therefore, the study proposes a method that combines a heuristic search algorithm and an improved ACO algorithm to improve the efficiency of obstacle avoidance for vessel navigation safety. Firstly, the standard ant colony optimisation algorithm is improved and applied to the study of obstacle avoidance paths for vessel navigation, and then the A* algorithm is effectively combined with the improved ACO algorithm to improve the performance of planning obstacle avoidance paths. Through simulation experiments and practical applications, the study verifies the capability of the obstacle avoidance planning model. The experimental results show that in simple environments, the hybrid algorithm reduces the path length by 3.8 and 5.5, and the number of iterations by 13.2 and 30.7 compared to Line-of-Sight and Particle Swarm Optimisation algorithms respectively. In moderately complex environments, the proposed algorithm reduces the average path length by 6.51 and 3.93 compared to Particle Swarm Optimisation and Line-of-Sight algorithms respectively. In complex environments, the proposed algorithm reduces the average path length by 15.7 and 12.4 compared to Particle Swarm Optimisation and Line-of-Sight algorithms, and reduces the number of iterations by 49 and 22.2, respectively. The study proposes a novel obstacle avoidance path planning method by effectively integrating the improved ant colony algorithm with the A* algorithm, which significantly improves the efficiency and accuracy of vessel navigation safety. The results show that the hybrid algorithm exhibits superior path planning capability in environments of varying complexity, and is able to quickly adapt to dynamically changing marine environments. This method not only provides a new solution for navigation safety, but also provides a theoretical basis and practical guidance for future autonomous navigation and decision-making of intelligent ships, which has important application value and promotion potential.