Evaluating connected and autonomous vehicles using a hardware-in-the-loop testbed and a living lab. (May 2019)

• Record Type: Journal Article
• Authors: Shao, Yunli; Mohd Zulkefli, Mohd Azrin; Sun, Zongxuan; Huang, Peter
• Is Part Of: Transportation research. Volume 102(2019)
• Page Start: 121
• Page End: 135
• Publication Date: 2019-05
• Subjects: Connected vehicle -- Hardware-in-the-loop test -- Engine testbed -- Fuel measurement -- Emissions measurement -- Autonomous Vehicle Evaluation; Transportation -- Periodicals; Transportation -- Technological innovations -- Periodicals; 388.011
• DOI: 10.1016/j.trc.2019.03.010
• Languages: English
• ISSNs: 0968-090X

Abstract

Highlights: A hardware-in-the-loop (HIL) testbed with an actual engine and a transient dynamometer. The HIL testbed emulates a virtual target vehicle with simulated vehicle dynamics. Performance of the virtual target vehicle matches an actual vehicle with 1% error. A living lab enables HIL testbed to interact with on-road vehicles in real-time. An effective and economical way to evaluate connected and autonomous vehicles. Abstract: Connected and autonomous vehicle (CAV) applications focusing on energy optimization have attracted a lot of attention recently. However, it is challenging to evaluate various optimizations and controls in real-world traffic scenarios due to safety and technical concerns. In light of this, our previous work has developed a hardware-in-the-loop (HIL) testbed with a laboratory powertrain research platform to evaluate CAV applications. An actual engine is loaded by a hydrostatic dynamometer whose loading torque is controlled in real-time by the simulated vehicle dynamics. The HIL testbed mimics the performance of a target vehicle and the dynamometer generates the same load as the target vehicle. In this work, the HIL testbed is further enhanced to match the performance of actual testing vehicles at the Federal Highway Administration (FHWA) and a living lab is developed to incorporate real traffic information. The same engine as the actual testing vehicles at the FHWA was installed and the vehicle models were calibrated using testing data from actual vehicles. The same roadway conditions (speed limit, the degree of road slope, etc.) were input into the testbed and the dynamometer generates the same load as an actual vehicle's engine is getting. The HIL testbed emulates the performance of an actual vehicle and both fuel consumption and emissions are measured by precise instruments in the lab. In eight testing scenarios, the results have shown that the performance of the HIL testbed matches very well with the actual testing vehicles with about 1% error. In addition, the living lab has enabled the HIL testbed to interact with real traffic and extended the testing capabilities of the HIL testbed. Two new testing capabilities have been demonstrated through two CAV applications. This is an exciting result since the HIL testbed could provide an effective and economical way for the testing of fuel consumption and emissions on various roadway conditions for CAVs and other types of vehicles.