UniPilot: Enabling GPS-Denied Autonomy Across Embodiments

Exploded view of the UniPilot module, with hardware, sensing, and compute components highlighted, alongside an image of the aerial platform that integrates the module in a process tank environment. Abstract-- This paper presents UniPilot, a compact hardware-software autonomy payload that can be integrated across diverse robot embodiments to enable autonomous operation in GPS-denied environments. The system integrates a multi-modal sensing suite including LiDAR, radar, vision, and inertial sensing for robust operation in conditions where uni-modal approaches may fail. UniPilot runs a complete autonomy software comprising multi-modal perception, exploration and inspection path planning, and learning-based navigation policies. The payload provides robust localization, mapping, planning, and safety and control capabilities in a single unit that can be deployed across a wide range of platforms. A large number of experiments are conducted across diverse environments and on a variety of robot platforms to validate the mapping, planning, and safe navigation capabilities enabled by the payload. I. INTRODUCTION With the progressing rise of autonomy, it becomes increasingly more important to be able to test a wide variety of complex methods and systems on easily re-creatable platforms. Especially of interest is to investigate and demonstrate a method's generality across different robot embodiments. Furthermore, the demand for autonomy in different robot actualization and in varied environments necessitates flexibility at the core of this design philosophy.