This paper addresses the Multi-Weeding Robot Task Assignment (MWRTA) problem, aiming to minimize the maximum completion time and residual herbicide in agricultural production. A Multi-Objective Teaching-Learning-Based Optimization (MOTLBO) algorithm is proposed to solve this problem. The MOTLBO algorithm includes an improved Nawaz Encore and Ham (NEH) heuristic and a maximum load-based heuristic for heuristic-based initialization, a dynamic grouping mechanism, and a multi-neighborhood-based local search strategy. The effectiveness of the MOTLBO algorithm is demonstrated through a comprehensive experiment comparing it with several state-of-the-art algorithms. The results show that the MOTLBO algorithm outperforms the other algorithms in terms of both the Inverse Generation Distance (IGD) and Hypervolume (HV) metrics, indicating its superior performance in solving the MWRTA problem.This paper addresses the Multi-Weeding Robot Task Assignment (MWRTA) problem, aiming to minimize the maximum completion time and residual herbicide in agricultural production. A Multi-Objective Teaching-Learning-Based Optimization (MOTLBO) algorithm is proposed to solve this problem. The MOTLBO algorithm includes an improved Nawaz Encore and Ham (NEH) heuristic and a maximum load-based heuristic for heuristic-based initialization, a dynamic grouping mechanism, and a multi-neighborhood-based local search strategy. The effectiveness of the MOTLBO algorithm is demonstrated through a comprehensive experiment comparing it with several state-of-the-art algorithms. The results show that the MOTLBO algorithm outperforms the other algorithms in terms of both the Inverse Generation Distance (IGD) and Hypervolume (HV) metrics, indicating its superior performance in solving the MWRTA problem.