Simulation-Based Acceptance Testing for Unmanned Ground Vehicles

摘要

Acceptance testing is considered a final stage of validation, and performing physical acceptance tests of an unmanned ground vehicle (UGV) can be expensive and time-consuming. Therefore, this paper discusses simulation-based acceptance testing and failure analysis for UGVs, which can be potentially used as a complementary tool in the design of efficient physical acceptance tests for reducing the testing time and cost for UGV products. In this paper, both dynamic and static simulation models are developed. A systematic statistical testing approach is presented to quantitatively assess when a simple static simulation model can be used to approximate a complex dynamic simulation. The results from the simulations show that a static simulation model can be appropriately used for determining the required joint motor torques of UGVs under slow operation speeds. One additional benefit of using a static simulation model is that it can explicitly identify the safe operation boundaries for preventing UGVs from joint torque saturation and rollover failures, thus improving the UGVs' operational reliability. On the other hand, the paper also shows that a dynamic simulation model is indeed needed to determine the maximum allowable moving speeds for UGVs to be safely operated on roads with various levels of roughness and bumpiness.