Bo Xue, Mengcheng Lin, and Xiangyu Wu
Robotic arm; polynomial interpolation; time-optimal trajectoryplanning; dung beetle optimiser algorithm; adaptive T-distribution
In the trajectory planning process of a robotic arm, it is essential to ensure that the driving device meets the actual load requirements. Consequently, the selection of joint velocities and accelerations tends to be relatively conservative. This cautious approach results in an extended duration required to complete a set of actions, ultimately hindering the full utilisation of the robotic arm’s continuity and stability based on its motion velocity and acceleration. To address the optimisation problem concerning the velocity and acceleration of various joints in a robotic arm, this paper focuses on the IRB2600 robotic arm as the subject of study. We introduce a 3-5-3 polynomial interpolation trajectory planning method within joint space and propose a time-optimal trajectory planning method based on a new multi-strategy improved DBO (NMSDBO) algorithm. This algorithm integrates the golden sine strategy, and incorporates adaptive T-distribution perturbation, which enhances the update of global optimal positions and improves both the accuracy and speed of the trajectory optimisation process. By comparing with WOA, HHO, SSA, and DBO algorithms in six benchmark functions, it is proved that the proposed NMDBO has good performance and robustness. The simulation results show that the time for each joint to complete the action is shortened from the initial set of 12 s to 5.5548 s, which achieves the effect of time optimisation and maintains the smooth and compliant motion of the manipulator.
Important Links:
Go Back
