On-Track Demonstration of Automated Eco-Driving Control for an Electric Vehicle

This paper presents the energy savings of an automated driving control applied to an electric vehicle based on the on-track testing results. The control is a universal speed planner that analytically solves the eco-driving optimal control problem, within a receding horizon framework and coupled with trajectory tracking lower-level controls. The automated eco-driving control can take advantage of signal phase and timing (SPaT) provided by approaching traffic lights via vehicle-to-infrastructure (V2I) communications. At each time step, the controller calculates the accelerator and brake pedal position (APP/BPP) based on the current state of the vehicle and the current and future information about the surrounding environment (e.g., speed limits, traffic light phase). The target vehicle is a Chevrolet Bolt, an electric vehicle, which is outfitted with a drive-by-wire (DBW) system that allows external APP/BPP to command the speed of the vehicle, while the operator remains in charge of the steering wheel. The DBW is connected to a rapid prototyping unit by dSpace. This unit includes: (1) real-time software that gathers all digital and analog sensors, as well as signals from the CAN bus; (2) a simple digital twin representation of the track; and (3) automated driving controls. The digital twin representation includes virtual stop signs, speed limits, and traffic lights. The digital twin can broadcast information about current and future road environment (e.g. SPaT) based on the actual position of the vehicle on the track, and correlate that to a position in the digital twin. The automated driving controls include eco-driving controls and an additional safety-focused control layer. The experiments include five road scenarios, and three control calibrations, and each combination is repeated three times. The road scenarios are all within 3.7 km in length, corresponding to one full loop around an oval track at the American Center for Mobility in Michigan, and feature various combinations of stop signs, traffic signals, and speed limits. The control calibrations correspond to a human-driver-like baseline, non-connected automated driving, and automated driving with V2I connectivity. Test-to-test variability is within 2%, thanks to careful thermal conditioning of the vehicle prior to tests. Functionality is verified and demonstrated: no excessive jerk and no violations of traffic laws occur. Energy savings of up to 7% are demonstrated in the no-connectivity case, and up to 22% in the V2I connectivity case. These tests demonstrate the real-world energy-saving potential of automated eco-driving controls.