Multi-objective optimization method for AMR dispatching in warehouse environments

To enhance warehouse logistics efficiency and reduce costs, the application of Autonomous Mobile Robots (AMRs) in warehouse logistics is becoming increasingly widespread. When a large number of orders arrive within a single cycle, determining how to effectively assign these transport tasks to multiple AMRs to minimize costs related to transportation and time has remained a challenging research problem. Addressing the multi-AMR task allocation issue, this paper proposes a multi-objective optimization algorithm based on the Non-Dominated Sorting Genetic Algorithm II, named the Residual Network-Assisted Non-Dominated Sorting Genetic Algorithm II (RA-NSGA-II). By leveraging Residual Network (ResNet) to predict the shortest path length for AMRs, the proposed approach efficiently completes the task allocation process, minimizing total transport path length and balancing path length loads for each AMR. Results indicate that this method is more efficient and achieves favorable multi-objective optimization outcomes.