Abstract: To address the problems of slow convergence, tortuous paths, and susceptibility to local optima in the path planning of coal mine robots caused by unstructured underground environments and dynamic obstacle interference, a path planning method for coal mine robots based on an Improved Dung Beetle Optimizer (IDBO) integrated with Artificial Potential Field (APF), namely the AIDBO algorithm, was proposed. To improve the insufficient population diversity of the traditional DBO algorithm, an IDBO algorithm was proposed. Singer chaotic mapping was introduced to initialize the population, which improved the uniformity of the initial solution distribution in complex roadways. A logarithmic spiral opposition-based learning strategy and a nonlinear adaptive contraction factor were integrated, which enhanced the ability of the algorithm to escape from local optima and jointly balanced global and local search. To solve the problem that the IDBO algorithm may still generate paths close to or even crossing obstacles in complex mine obstacle scenarios, a fitness–obstacle dual-driven gain-adaptive strategy was constructed. The APF was embedded into the IDBO algorithm as a post-processing operator, and the potential field mechanism was used to perform secondary smoothing and local fine-tuning of the global path. Simulation results showed that in a 30×30 static grid map, the shortest path length planned by the AIDBO algorithm was 45.053 7 cm, which was reduced by 15.4% compared with the IDBO algorithm and by 22.0% compared with the DBO algorithm. In a 40×40 dynamic grid map, the shortest path length planned by the AIDBO algorithm was 58.526 1 cm, which was reduced by 20.3% compared with the IDBO algorithm and by 28.2% compared with the DBO algorithm. Compared with five other algorithms such as particle swarm optimization and sparrow search algorithm, the AIDBO algorithm exhibited faster convergence speed and achieved the shortest planned path length, and it maintained excellent robustness and computational efficiency in complex dynamic scenarios.