Global trajectory optimization of mobile robot based on Minimum Snap

Since the path planned by the traditional A * algorithm consists of a series of discrete spatial points and does not contain time and control quantities, it cannot be directly used for trajectory tracking of mobile robots. To address this problem, this paper uses the Minimum Snap method to transform the traditional path planning algorithm and the planned path into a continuous smooth trajectory containing both time, velocity and acceleration information. The trajectory is optimized by introducing the trajectory shape constraint and velocity and acceleration constraint, so as to generate a global trajectory suitable for the path tracking of mobile robots, while the generated trajectory differs too much from the original path shape and the robot motion trajectory is not smooth due to the robot's own physical constraints are not considered. Finally, the effectiveness of the proposed method is verified by simulation experiments, and the method is applied to a robot autonomous navigation system under ROS (Robot Operating System) to confirm the practicality and reliability of the method.