Ecological Cooperative Adaptive Cruise Control for Autonomous Electric Vehicles

This thesis develops an ecological cooperative adaptive cruise control, which exploits look ahead information coming from a preceding vehicle in order to minimize energy consumption. The controller enables substantial fuel savings, combining an optimal eco-driving and an adaptive cruise control that exploits the reduction of aerodynamic resistance. The proposed control approach is tailored for electric vehicles as it leverages a realistic powertrain model, validated with experiments in the literature. Different optimization techniques are analyzed and compared to solve the optimal control problem offline. For real-time implementation, a nonlinear model predictive control framework is proposed. Look-ahead information is simulated using experimental data collected driving in the Silicon Valley around the city of San Francisco. Simulations in real-world driving conditions suggest substantial energy savings when the proposed Eco-CACC is compared to a standard adaptive cruise controller. Finally, hardware-in the-loop tests have been conducted to show the effectiveness of the approach for vehicle implementations.