Cheerleading events are flourishing in China, the level of competition is rising, the number of competition groups and programs is increasing, the competition is becoming more and more intense, and the innovative research on formation design is an inevitable demand for the development trend of cheerleading. The study designed a multi-objective path planning model based on the intensity of willingness and consultation strategy, so that college cheerleading can avoid conflicts and reach the goal point of cheerleaders in the complex environment. Then an improved multi objective particle swarm algorithm (MOPSO-CA) based on meta cellular automata is proposed and applied to college cheerleading formations to realize the design of college cheerleading formations. The simulation results show that the MOPSO-CA algorithm can re-select the optimal movement direction angle according to the real-time positions of the moving obstacles and moving targets, which illustrates the effectiveness of the algorithm. Secondly the feasibility of the formation design conditions are suggested as: keeping the originality of the movement, the use of the moving route of the formation and the space of the venue, and the type of formation change.