A Virtual Driveline Concept to Maximize Mobility Performance of Autonomous Electric Vehicles 2020-01-0746
In-wheel electric motors open up new technical prospects to radically enhance mobility performance of autonomous electric vehicles with four or more driving wheels. The flexibility and agility of the e-motor torque delivered individually to each wheel creates conditions for significant improvements of mobility, providing agile maneuvers while maintaining stability, and increasing energy efficiency of vehicles.
However, the driving of individual wheels, when the wheels are not connected mechanically by a driveline system, does not mean that the e-motor drives are independent from each other and do not impact each other’s actions. Indeed, loaded with individually supplied different torques, the wheels will be loaded with different longitudinal forces that act from the vehicle frame to the wheel axis and, thus, the tires will generate different traction forces in the patches with terrain and develop different tire slippages. Thus, the absence of driveline systems, which physically connect the wheels and control the wheel torque split in vehicles with mechanical powertrains, requires conceptually new approaches to coordinate the torque distribution to the driving wheels in electric vehicles.
With regard to the above-formulated problem, this paper solves two technical problems. First, the concept of the virtual driveline system (VDS) is proposed to emulate a mechanical driveline system that virtually connects the driveshafts of the e-motors and provides a coordinated torque management of the driving wheels. Arranged as a computer code, VDS simulates the split power between the driving wheels. Conceptually, VDS is founded on the generalized tire and vehicle parameters, including the generalized rolling radii of the two front wheels, two rear wheels, and the vehicle. Two generalized slippages of the two tires at the front and two tires at the rear, and the generalized slippage of the vehicle are introduced and then utilized to determine the virtual gear ratios from a virtual transfer case to each wheel. The virtual gear ratios serve as the voltage signals to the electric motors to change the torques.
As the second technical problem solved in the paper, a new velocity-based mobility performance index is introduced as the ratio of the actual velocity of a vehicle with the individual wheel management to the theoretical velocity (i.e., the velocity at zero tire slip) of the same vehicle, but equipped with a mechanical driveline, whose gear ratios to the driving wheels are not controlled. Using the velocity-based mobility performance index as the objective function, a math maximization problem of vehicle mobility is formulated and solved in the paper. The optimal virtual gear ratios, i.e., the control voltage of the e-motors, are determined to provide the maximal possible mobility in a given terrain condition.
Computer simulations of a 4x4 tactical vehicle with the gross mass of 9.5 ton in stochastic soft soil conditions demonstrated a 17% increase in the mean values of the velocity-based mobility performance index when the vehicle is electrically driven by the optimal virtual gear ratios as compared to the mechanical driveline system with non-controlled differentials.
Vladimir Vantsevich, David Gorsich, Jesse R. Paldan, Michael Letherwood
University of Alabama at Birmingham, U.S. Army Ground Vehicle Systems Center, Alion Science & Technology