Predictive Energy Optimization for Connected and Automated HEVs

Connected and automated vehicles (CAVs) have the potential to eliminate road vehicle collisions and other traffic incidences. Whilst the main motivation for the introduction of vehicular communication systems is to improve safety, they also provide opportunities to reduce CO₂ and other harmful pollutant emissions as well as transportation energy costs. Vehicle communication link with other automobiles and Intelligent Transportation Systems (ITS), when combined with the use of on-board high definition navigation maps, enable the vehicle control systems to optimize their operation and streamline traffic flow. This paper presents the development and evaluation of proof of concept control algorithms which optimize the vehicle’s propulsive energy consumption. Consideration is also given to journey time and other drivability and autonomous driving attributes and constraints. A two-stage optimization approach is used to optimize the ego-vehicle speed trajectory and powertrain state of a Plugin Hybrid Electric Vehicle (PHEV), within a receding prediction horizon. The control algorithm is based on the Dynamic Programming (DP) optimization method. The algorithm calculates upper and lower speed constraints that have to be respected by the ego-vehicle in the optimization window. The speed constraints are updated dynamically. They account for static road features and time-varying events as dictated by the road traffic situation and traffic light phasing. The developed functions could be deployed as part of an Economy (ECO) setting of an autonomous feature which holistically delivers a set of functionalities equivalent to the combined operation of Adaptive Cruise Control (ACC), Intelligent Cruise Control (ICC), Green Light Optimized Speed Advisory (GLOSA), and Traffic Jam Assist (TJA). Alternatively, the function could be used to coach the driver to follow a target speed, and indicate when to lift off the accelerator pedal or when to change gear.