Eco-Approach and Departure Algorithm for Connected and Automated PHEVs: Simulation and in-Vehicle Results

Plug-in Hybrid Electric Vehicles (PHEVs) combine the benefits of electric propulsion and storage with the extended range of conventional internal combustion engines to reduce fuel consumption and greenhouse gas emissions. However, optimizing the efficiency of PHEVs in real-world driving conditions remains a challenge due to the uncertainties of environmental and driving conditions. Connectivity and automation technologies can offer a unique opportunity to enhance the efficiency of PHEVs by enabling real-time interaction with surrounding vehicles and infrastructure. By leveraging these technologies, significant reductions in energy consumption for PHEVs can be achieved. However, most existing works primarily rely on simulation-based analyses to evaluate energy savings offered by connected and automated PHEVs. This study advances the understanding of the energy-saving potential of connected and automated PHEVs by incorporating experimental validation alongside simulation-based analyses. The activity relies upon an Eco-Approach and Departure (Eco-AND) algorithm, previously developed as a speed planning and energy management strategy for PHEVs to optimize energy savings by leveraging route information from navigation systems, GPS, and Vehicle-to-Infrastructure (V2I) communication. The algorithm minimizes the trade-off between fuel consumption and PHEV travel time over a given itinerary. In this study, the algorithm is implemented as a receding horizon nonlinear optimal controller for spatiotemporal trajectory optimization, incorporating real-time traffic conditions and signal phase and timing (SPaT) information. A Monte Carlo simulation framework is first created to quantify the benefits of the proposed Eco-AND strategy in real-world driving scenarios. This framework evaluates the impact of SPaT information, traffic variability, and driving aggressiveness on equivalent fuel consumption and travel time. Then, experimental results obtained via in-vehicle testing performed on a test track are analyzed to confirm and verify the initial findings of a subset of the simulated scenarios, enabling a direct comparison between simulated predictions and real-world performance.