Mobile Robot Path Planning in Complex Road Conditions Based on Hybrid Ant Colony Algorithm

To address the challenges of intelligent path planning for mobile robots in complex road environments, this paper proposes an innovative Hybrid Ant Colony Algorithm (HACO). First, a novel complex road environment model is constructed, where the map is represented as a neural network composed of multiple interconnected neurons, and the variation in neuronal activity is utilized to characterize complex road conditions. Then, a goal-enhanced path decision model is introduced, which leverages neural network path activity to guide the robot through target-driven stimulation while avoiding hazardous road segments through inhibition effects from obstacles and the surrounding complex environment. Finally, an improved pheromone update mechanism is implemented in the ant colony algorithm, enhancing both global and local optimization capabilities. Experimental results demonstrate that compared to traditional genetic algorithms and ant colony algorithms, the proposed HACO reduces the number of turns by up to 11, shortens the path length by up to 9.37%, and significantly improves convergence speed. Additionally, the feasibility of the algorithm is validated in dynamic environments. The proposed method effectively enhances the efficiency and reliability of mobile robot path planning in complex road environments, showcasing its potential applications in the field of intelligent control.