The Virtual Autonomous Navigation Environment: High Fidelity Simulations of Sensor, Environment, and Terramechanics for Robotics

Abstract

The Virtual Autonomous Navigation Environment (VANE) is a modeling and simulation tool that is being developed by the US Army Engineer Research and Development Center (ERDC) to support the design, development, and testing of autonomy algorithms for intelligent ground vehicles. The VANE is an open architecture, modular simulation that is compatible with desktop and parallel computing. It consists of a set of independent models for sensors, vehicle dynamics, and terramechanics. The VANE uses high-fidelity environment geometry and ray-tracing to simulate GPS, Light Detection and Ranging (LIDAR), and camera sensors. Inertial sensors are simulated using the dynamics engine, which also interfaces with the Vehicle Terrain Interaction (VTI). The VTI is a validated mobility model that gives performance predictions for both hard and soft soils for tracked and wheeled vehicles. These models interact through a run time interface in a high performance distributed computing environment and can be used to simulate environmental effects on sensors and mobility.