Terrain aware tactical motion planning and control algorithms for off-road UGVs in GNSS denied hostile environments

Abstract

This dissertation introduces an advanced framework for terrain-aware tactical motion planning and control of off-road unmanned ground vehicles operating in environments where global navigation satellite systems are unavailable due to adversarial interference or structural constraints. The research focuses on enabling autonomous navigation in uncharted, hostile terrains by developing a hierarchical autonomy stack that seamlessly integrates navigation, path planning, trajectory planning, and control functionalities. The navigation system employs onboard vision-based and inertial sensors to construct real-time environmental representations, utilizing geometric segmentation techniques such as random sample consensus and inverse ray tracing to differentiate traversable surfaces from obstacles. These representations, encompassing occupancy grids and topographic profiles, account for terrain geometry and surface characteristics, providing a foundation for subsequent planning stages. The path planning module leverages heuristic-driven graph search strategies to compute waypoint sequences that optimize for tactical concealment while ensuring efficient progression toward a goal. A novel stealth-inducing mechanism biases paths to exploit obstacle proximity for cover, and a dual-mode interaction paradigm distinguishes between protective shelters and hazardous entities, enhancing strategic navigation in contested settings. The trajectory planning module transforms these waypoints into smooth, time-parameterized trajectories through cubic polynomial spline interpolation, incorporating terrain-adaptive orientation via rotation-minimizing frames to maintain kinematic consistency over uneven landscapes. A tactical velocity modulation scheme adjusts motion dynamics based on terrain elevation and obstacle proximity, while collision avoidance is achieved through the generation of safe navigation corridors using geometric constructs. The control system ensures robust trajectory execution, compensating for environmental uncertainties. Field deployment on a robotic platform in a controlled indoor environment validates the framework's capability to map unknown terrains, generate stealth-aware paths, and produce feasible trajectories under GNSS-denied conditions. The research advances autonomous off-road navigation by offering scalable algorithms that enhance tactical decision-making, with potential applications in military reconnaissance, disaster response, and extraterrestrial exploration.