Robot Path Planning Based on Ant Colony Optimization Algorithm for Environments with Obstacles

In modern industrial processes, robotic equipment is widely used, and one of the most pressing problems is to have to have navigation available for mobile robots. In this paper, the ant algorithm for laying and optimizing the robots paths in 2-dimensional environments with obstacles, is described and shown on construction site examples. The most important requirement is to be able to plan the shortest or permissible robot navigation route in such a complex environment with obstacles. It is well known that one of the most effective solutions to resolve such optimization problems of route seals is provided by the ant colony optimization (ACO) algorithm. The exploratory nature of the ant colony behaviour requires a classical partition of the search space, which is incomparably smaller, when compared to the obstacles fragments, as considered within this paper. The ant’s agents use the traditional logic of selecting the transition from fragment to fragment: the memory of the most popular routes based on pheromone are investigated, and formulated within the task elements, adopting appropriate tactics and situational awareness, and based on the random decisions. In addition, the new elements of the decision-making tactics are formulated for each task. For example, “feeling” of targeted routes by laying points is added to the algorithm. The natural analogue of this mechanism is similar to sensing the odors by the mustaches of real ants. The special software tool “Path Planning Optimization with Obstacle Avoidances by Ant Algorithm” is designed as the research test bed. A comprehensive study of the proposed algorithm, which shows superior performance, is done by utilizing the developed software. The examples, of the construction site with different complexity, are provided to explain the finding of the suboptimal routes for the specially designed test tracks, with defined obstacles in the simulated construction site landscape. The analysis of the results confirms the relevance and effectiveness of the developed software, which is based on the ant algorithm for the robot path planning, and validated for the environments containing complex obstacles.