Abstract:
The lack of structure in off-road terrains makes off-road operations of automated platforms difficult. The difficulty arises from uncertainty in the optimality and safety of the actions (e.g., planning and control) taken by the automated platform. When multiple automated platforms are required to act in a coordinated manner (e.g., a convoy) in complex cluttered off-road environments, this uncertainty can further affect the scheduling and routing of all agents. While prior works consider the challenges with off-road driving and convoy operations separately, this thesis proposal investigates the development of a unified framework of allocation, planning, and control policies with a focus on navigation- based tasks for automated convoying in off-road environments. The proposed work in this document builds upon developed control strategies to override vegetation in off-road terrains and to coordinate the actions of convoying automated platforms. An instance of the vehicle routing problem with multiple synchronization constraints known as the Vehicle Routing Problem with Restricted Clustering (VRPRC) is introduced to capture the unique constraints of the control of the automated convoy and then extended to address environmental dynamism. The designed control strategies enable further development of allocation and planning problems with environmental contacts, which is a central problem to deploying automated convoying systems in off-road environments and is the focus of this thesis.

Thesis Committee Members:
Sebastian Scherer, Co-chair
Matthew Travers, Co-chair
Tuomas Sandholm
Ali-akbar Agha-mohammadi, Field AI

More Information