At present, the coal mine rescue robot is without the power supply in the hazard and complex coal mine and the explosive-proof batteries unit influences dynamic performance very seriously. It is the only method to solve the two problems above by power matching and optimization. Therefore, the power matching design method based on multi-objective particle swarm optimization (PSO) is proposed. The optimization goal and constraint condition for power matching are confirmed by the analysis of the robot dynamic performance requirement. And the model of the robot power matching is established and the decision variables of multi-objective optimization are ensured, which are based on the tracked vehicle dynamic and the influence for the dynamic performance by explosive-proof batteries unit. Finally, it is calculated which the reasonable value range of the parameters of power matching. The test indicates that the mass of the coal mine rescue robot has reduced by 24.36% and the time of endurance is double. As well as, it is supported the multi-objective particle swarm optimization could effectively and rapidly solve the power matching of the coal mine rescue robot.