In an era where unmanned ground vehicles (UGVs) are becoming indispensable in both civilian and military operations, researchers have turned their attention to optimizing their deployment in complex environments. A recent study, titled “Heuristics for Multi-Vehicle Routing Problem Considering Human-Robot Interactions,” addresses the intricate challenges of integrating UGVs with manned ground vehicles (MGVs) in a leader-follower framework. This research, conducted by Venkata Sirimuvva Chirala, Kaarthik Sundar, Saravanan Venkatachalam, Jonathon M. Smereka, and Sam Kassoumeh, offers a novel approach to enhancing mission efficiency while minimizing operational costs.
The study focuses on a multi-objective, multiple-vehicle routing problem where teams of MGVs and UGVs collaborate to execute missions with varying requirements. The complexity of this problem is compounded by the need to consider human-robot interactions (HRI), which significantly impact the effectiveness of team dynamics. HRI studies reveal that managing a team of UGVs by an MGV incurs specific costs, necessitating a careful balance between mission requirements and operational efficiency.
The researchers have developed a comprehensive model to address these challenges. They first formulated the problem as a mixed-integer linear program (MILP), which can be solved to optimality using commercial solvers for small-sized instances. However, recognizing the limitations of this approach for larger, more complex scenarios, they also proposed a variable neighborhood search algorithm. This heuristic method is designed to compute near-optimal solutions, thereby tackling the combinatorial multi-objective routing optimization problem with greater efficiency.
The study’s primary objective is to compute feasible paths, replenishments, team compositions, and the number of MGV-UGV teams deployed. The goal is to meet mission requirements while minimizing path, replenishment, HRI, and team deployment costs. The researchers’ innovative approach ensures that the unique capabilities and limitations of both MGVs and UGVs are considered, leading to more effective mission execution.
To validate their methods, the researchers conducted computational experiments that demonstrate the effectiveness of the proposed algorithms. These experiments underscore the practical applicability of the solutions in real-world scenarios, where mission success hinges on optimal resource allocation and efficient team coordination.
The implications of this research extend beyond theoretical advancements. In the defence and security sector, the ability to optimize multi-vehicle routing problems can significantly enhance mission capabilities. Whether in underground mining, nuclear plant operations, planetary exploration, or intelligence, surveillance, and reconnaissance (ISR) missions, the integration of MGVs and UGVs offers a strategic advantage. Moreover, the study’s findings can inform the development of future military strategies, ensuring that human-robot teams are deployed in the most effective and cost-efficient manner possible.
As the defence landscape continues to evolve, the need for advanced algorithms that can handle complex, multi-objective problems becomes increasingly critical. This research provides a robust framework for addressing these challenges, paving the way for more sophisticated and efficient military operations. By leveraging the strengths of both manned and unmanned systems, defence planners can achieve greater operational flexibility and mission success, ultimately enhancing the safety and security of personnel and assets. Read the original research paper here.