Route the Gap: A Semi-Autonomous Wet Gap Crossing
Abstract
An interdisciplinary team composed of undergraduate students from the University of Illinois at Chicago, College of Engineering intends to compete in the ARL HBCU/MI Undergraduate Student Design Competition. The competition simulates the wet gap crossing problem, targeting the minimum possible time to build and cross a water gap using floating bridge sections while neutralizing enemy threats. A ground vehicle, tank, and boat, are simulated by NVIDIA Kaya analogues, which are programmed to be principally autonomous. The ground vehicle and boat place bridge sections, while the tank neutralizes two threats that intend to disrupt the communications and control of the vehicles. A stereoscopic depth camera, in addition to an inertial measurement unit, localizes the vehicles in space and allows for object detection. We present two strategies for building the bridge; the first strategy moves alternating bridge sections by coordinating ground vehicle and boat movements, such that the section is pinched between the vehicles while being moved into position. The second strategy builds a temporary bridge, moving from right to left or vice-versa, and each section is placed further cross-stream. After placement of a new section, the entire bridge is shifted up or down-stream. The vehicles utilize Wi-Fi to coordinate activities and update the map. Simulation of the strategies in NVIDIA Isaac Sim determines the most effective approach. Approaches for mitigating risks relating to orientation of the bridge sections, interruption of communications, and conflicts relating to update of the map in memory, are presented.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 24, 2020
- Source ID
- W911NF2010042
Entities
People
- Jonathan Komperda
Organizations
- Army Contracting Command
- Office of the Secretary of Defense
- University of Illinois at Chicago