Multi-Objective Trajectory Planning with Dual-Encoder

TL;DR

This paper introduces a dual-encoder based transformer model to establish a good preliminary trajectory and refined through sequential quadratic programming to improve its optimality and robustness, and shrinks the optimality gap.

Abstract

Time-jerk optimal trajectory planning is crucial in advancing robotic arms' performance in dynamic tasks. Traditional methods rely on solving complex nonlinear programming problems, bringing significant delays in generating optimized trajectories. In this paper, we propose a two-stage approach to accelerate time-jerk optimal trajectory planning. Firstly, we introduce a dual-encoder based transformer model to establish a good preliminary trajectory. This trajectory is subsequently refined through sequential quadratic programming to improve its optimality and robustness. Our approach outperforms the state-of-the-art by up to 79.72\% in reducing trajectory planning time. Compared with existing methods, our method shrinks the optimality gap with the objective function value decreasing by up to 29.9\%.

Figures (7)

• Figure 1: We show a 4DOFs robot manipulator. A joint space trajectory composes waypoints for each joint and an interpolation function passes through the waypoints.
• Figure 2: Dual-Encoder based Transformer architecture.
• Figure 3: Comparison of trajectories generated by our dual-encoder model (dashed) to the ground truth generated by 5th-order BSpline (solid).
• Figure 4: Convergence trends of SQP, our method, IPTP, TOTG, NSGA-II, LSTM with various numbers of waypoints over time.
• Figure 5: Time-jerk optimal trajectory data collection with MoveIt2.