With the progress of industrial automation, industrial robots play an increasingly important role in the upgrading of the manufacturing industry. Trajectory planning is the key technology of precise motion control of industrial robots. At present, most industrial robots use the method of teaching programming for trajectory planning, which has the problems of low efficiency, long time-consuming and poor accuracy. Simulation programming overcomes many shortcomings of teaching programming. It uses information technology to program, debug and run in simulation scenarios, which reduces time costs and improves planning efficiency and accuracy. It has attracted more and more attention from manufacturing enterprises. In this paper, the trajectory planning process based on MATLAB platform is studied by taking GSK RB08A3 6-axis industrial robot as the research object. Firstly, the coordinate system of each link of RB08A3 robot is established according to the standard DH parameter method, and the mathematical model is built according to the parameters of the robot. The forward and inverse kinematics verification is completed by using the functions in the Robotics Toolbox of MATLAB, which further proves the correctness of the modeling. Then, the multi-trajectory planning problem of industrial robots in Cartesian space is studied. A continuous multi-path smooth transition algorithm is designed based on the improvement of S-shaped speed curve, and the simulation experiment is carried out on the MATLAB platform compared with the traditional S-shaped acceleration and deceleration algorithm with zero head and end speed. The results show that the execution efficiency of the algorithm is greatly improved compared with the traditional S-shaped speed curve trajectory planning, and the effect of smooth transition of adjacent trajectories is achieved, and the mechanical impact is also reduced. Finally, aiming at the problems of working efficiency, energy consumption and flexible impact during start-stop of industrial robots, a time-energy optimal 5-3-5 piecewise interpolation joint space trajectory planning method is proposed. The simulation results show that the optimized piecewise interpolation trajectory can effectively avoid the flexible impact during start-stop, greatly improve the working efficiency of industrial robots, greatly reduce energy consumption, and achieve the comprehensive optimization of time-energy consumption, which verifies the effectiveness of the trajectory planning method.