Distributed Sliding Mode Control For Heavy-duty Truck Platooning

The worldwide demand for freight transport is expected to keep increasing in the future.However,road capacities are limited and demanding governmental emission regulations are challenging the global transportation sector.With rapidly improving information technology,platooning of automated heavy-duty trucks travelling together as a collection by communicating with each other can help to overcome these challenges.Platooning bears the potential not only to improve road traffic utility by reducing congestions,enhancing safety and increasing road capacity,but also to significantly decrease fuel consumption due to reduced aerodynamic drag at short intervehicle distance.To address these potentials,this study presents a distributed sliding mode control(SMC)approach ensuring accurate tracking of vehicles within the platoon.After setting up a simulation model for truck platooning,the distributed sliding mode controller is designed with the focus on the platoon members to track a desired inter-vehicle distance.The model is then used to investigate different driving scenarios and to evaluate the performance of the platoon controller according to predefined indices.(1)The simulation model for a homogeneous platoon is built including one leading vehicle and ten following vehicles,each of which represents the vehicle longitudinal dynamics of an ISUZU heavy-duty truck.The model is extended to account for the reduction of aerodynamic drag each vehicle faces depending on inter-vehicle distance and vehicle position.(2)Active inter-vehicle communication through both radar-based sensors and vehicle-to-vehicle information exchange is an essential part in platooning operation.In the simulation model,different information flow topologies are included in order to describe how the platoon members communicate with each other.The topologies are mathematically described by directed graphs and thereby define from which neighbors each vehicle receives information on the momentary vehicle velocity and torque error.(3)The designed platoon controller is divided into two parts,which are the upperlevel and the lower-level controller.The upper-level is designed as a distributed sliding mode controller based on a first-order transfer function to represent the powertrain dynamics.It consists of a sliding surface and a reaching law according to the information flow topology and generates the desired drive torque for each vehicle.In the lower-level controller,an inverse vehicle model is implemented to compensate the nonlinearities in the vehicle dynamics,and the desired throttle position and brake system input are generated as the controller output.(4)Simulations are conducted for the cases where the leading vehicle is either driving at constant speed or following the standard HHDDT drive cycle,while the remaining platoon members are following in accordance to the controller output.The controller performance depending on six different topologies is evaluated with respect to tracking capabilities,traffic smoothness,safety and fuel economy.It is shown that tracking is generally improved by adding information flow from additional preceding vehicles.In addition,the influence of the fuel reduction potential depending on desired inter-vehicle distance is investigated.As it is shown,the average fuel consumption at a desired inter-vehicle distance of 5 m can be reduced by 21.63 % at constant speed of 25 m/s through reduction of aerodynamic drag,and by 13.27 % in the HHDDT drive cycle.