Fuel-Efficient Control of Merging Maneuvers for Heavy-Duty Vehicle Platooning

Abstract: With the rapid increase in the research of promising technologies such as vehicle platooning, which are trying to find more fuel-efficient ways of operating our transportation systems, complex questions requiring creative solutions arise. Regarded as an up-and-coming technology that is expected to be highly relevant and widely applied in the near future, vehicle platooning allows the smart manipulation of the traffic by grouping two or more vehicles into a single convoy and coordinating its behavior, all the while guaranteeing a safe and fuel-efficient freight operation. This thesis tackles the specific problem in which two platoons of heavyduty vehicles (HDVs), coming from different highways, would like to be merged into a single one and remain driving as a larger platoon using the least amount of fuel possible. This can be beneficial if the HDVs need to travel a long distance after the merging point and fuel savings of up to 8% can be achieved by platooning along this trajectory. The focus lies on the calculation of the optimal input policies that minimize the fuel consumption of both platoons while completing the merging maneuver. Some theoretical background on optimal control and the minimum principle is introduced and it is consequently applied on a nonlinear model of the HDVs dynamics. The culmination of this approach is the solution of the inputconstrained, free end-time optimization problem. Due to the non-linearity of the model and the conditions set by the minimum principle numerical integration techniques, such as the shooting method, are also presented. In a final stage, this optimization tool is used together with shrinking horizon model predictive control to create a controller that is capable of accounting for disturbances, speed constraints and model uncertainties. The final result of this thesis is a control algorithm that can merge two platoons of any length, even in the presence of undesirable disturbances by correcting the optimal policies accordingly, in the most fuel-efficient manner.