Research on optimized time-synchronous online trajectory generation method for a robot arm

Abstract

Based on sensor input, a robot arm should dynamically adjust its trajectory while maintaining stability to react to a sudden change in the target point in an unknown environment. To solve this problem, in this study, a time-optimized online trajectory generation (OTG) method is proposed using an S-curve velocity profile, which can generate trajectories in response to external sensor signals. The generated trajectory has characteristics that can guarantee the synchronization of multijoints according to an arbitrary initial state and a desired target state under velocity, acceleration, and jerk constraints. For multijoints time synchronization, two different characteristics are considered according to different application scenarios: minimum velocity or peak acceleration, which correspond to two sub-methods. The first one can be used to calculate with a minimum velocity peak, which can quickly adjust the trajectory according to the signal feedback. The second can be used to calculate the minimization of the acceleration peak, which can reduce the vibration of the robot arm due to a change in the motion. Compared with other OTG methods, the second proposed sub-method can effectively reduce the acceleration peak of the planned motion with the same synchronization time and parameters. Additionally, both sub-methods have the advantage of a rapid calculation and can generate time synchronization motion trajectories for 6 axes in 0.21 ms on a personal computer, fully satisfying the requirements of online motion planning. Finally, the effectiveness of the algorithm is verified by simulations and experiments with a lab-developed robot arm.