A Scaled Automotive Platform for Validation and Testing of Perception and Control Algorithms for Unmanned Ground Vehicles Operating Under Extreme Driving Conditions

Abstract

This report summarizes efforts to create a new high-speed scaled autonomous vehicle and accompanying outdoor test facility at Georgia Tech via DURIP Award W911NF-12-1-0377 to test and validate newly developed perception and control algorithms for autonomous unmanned ground robotic vehicles (UGVs) operating in uncertain and unstructured environments, and under extreme driving conditions (i.e., off-road, at high speed, while skidding/slipping, etc). Two vehicles were constructed, each approximately 1 m long, 0.6 m wide, 0.4 m high, weighing 20.5 kg, with a top speed of 90 kph. The platform can drive itself fully autonomously using only on-board sensing, computing, and power. This hardware complements the theoretical work of the PI and his co-workers in the area of integrated, neuro-inspired perception and control of autonomous, aggressively driven vehicles, developed under MURI ARO award Neuro-Inspired Adaptive Perception and Control for Agile Mobility of Autonomous Vehicles in Uncertain and Hostile Environments (ARO award no. FA9550-04-1-0135). This research will enable radically new capabilities of autonomous and semi-autonomous ground vehicles in the battlefield. It will enhance manyfold the maneuverability and mission impact of these vehicles by allowing them to move at high speed, over rough terrain and while operating in hostile environments. While smaller than many other research-focused autonomous ground vehicles, the platform offers a cost-effective, fast, agile, and safe alternative to operating full-sized autonomous vehicles while retaining realistic vehicle dynamics that the smaller toy platforms lack. The sensor package, hardware, and computing capabilities offer a large performance improvement over traditional scaled autonomous vehicles. Several graduate and undergraduate students use the robots and test site for high-speed control, vision, and autonomy research.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 01, 2015
Accession Number
AD1224618

Entities

People

  • Panagiotis Tsiotras

Organizations

  • Georgia Tech

Tags

Readers

  • Research Science/Academic Research
  • Systems Analysis and Design
  • Unmanned Aerial System (UAS) Autonomous Capabilities and Mission Reconnaissance.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • Autonomy
  • Autonomy - Autonomous System Control