Abstract: Aiming at the difficulty of tuning the parameters of the linear active disturbance rejection controller for the lumbar exoskeleton robot, an improved particle swarm optimization algorithm (PSO) based on the beetle antennae search is proposed. First, this study establishes a lumbar exoskeleton robot model, then designs the linear active disturbance rejection controller, and finally introduces the improved PSO to optimize its parameters. The algorithm initializes the population through chaos to improve the execution efficiency of particles and adopts nonlinear strategies to adjust inertia factors and learning factors to strengthen the search ability of particles. The long beetle search algorithm combined with PSO and using the adaptive weighting factor is introduced so that particles can judge the surrounding environment and avoid particles from falling into a local optimum. The simulation experiments are performed through six test functions and the establishment of system evaluation indicators. The results show that the proposed algorithm has good convergence accuracy, and the optimized controller has good control performance.