This paper presents a novel real-time tire-road friction coefficient estimation method that is independent of vehicle longitudinal motion for ground vehicles with separable control of the front and rear wheels. The tire-road friction coefficient information is of critical importance for vehicle dynamic control systems and intelligent autonomous vehicle applications. In this paper, the vehicle longitudinal-motion-independent tire-road friction coefficient estimation method consists of three main components: 1) an observer to estimate the internal state of a dynamic LuGre tire model; 2) an adaptive control law with a parameter projection mechanism to track the desired vehicle longitudinal motion in the presence of tire-road friction coefficient uncertainties and actively injected braking excitation signals; and 3) a recursive least square estimator that is independent of the control law, to estimate the tire-road friction coefficient in real time. Simulation results based on a high-fidelity CarSim full-vehicle model show that the system can reliably estimate the tire-road friction coefficient independent of vehicle longitudinal motion.
1 July 2014 in IEEE Transactions on Control Systems Technology
A hierarchical control structure, consisting of a high-level dynamic sliding mode control (SMC) and a low-level adaptive energy-efficient control allocation (A-EECA) scheme, is presented to track the planar motions of an electric ground vehicle with four in-wheel motors while achieving the optimal energy consumption. By explicitly incorporating the efficiency functions and input constraints of in-wheel motors in the low-level A-EECA design, virtual control signals from the high-level dynamic SMC are distributed to four actuators with an adaptive convergence to the energy-optimal operating points. Taking allocation errors between the virtual control efforts and the real actuation realizations as inputs, the input-to-state stability of the overall feedback system is proved. Both simulation and experimental results in different maneuvers are demonstrated to validate the control design. The high-level vehicle tracking performance, low-level torque distributions, and total energy consumptions in the test maneuvers are compared between the proposed A-EECA and a standard pseudoinverse control allocation without considering the power optimization.
12 August 2011 in IEEE Transactions on Control Systems Technology
This paper presents a fast and global optimization algorithm for an energy-efficient control allocation (CA) scheme, which was proposed for improving the operational energy efficiency of over-actuated systems. For a class of realistic actuator power and efficiency functions, a Karush-Kuhn-Tucker (KKT)-based algorithm was devised to find all the local optimal solutions, and consequently the global minimum through a further simple comparison among all the realistic local minima and boundary values for such a non-convex optimization problem. This KKT-based algorithm is also independent on the selections of initial conditions by transferring the standard nonlinear optimization problem into classical eigenvalue problems. Numerical examples for electric vehicles with in-wheel motors were utilized to validate the effectiveness of the proposed global optimization algorithm. Simulation results, based on the parameters of an electric ground vehicle actuated by in-wheel motors (whose energy efficiencies were experimentally calibrated), showed that the proposed global optimization algorithm was at least 20 times faster than the classical active-set optimization method, while achieving better control allocation results for system energy saving.
14 February 2012 in IEEE Transactions on Vehicular Technology
This paper discusses the design and evaluation of electric differentials (ED) for overactuated electric ground vehicles (EGVs) with four independent in-wheel motors. Three patterns of ED, namely, 1) front ED, 2) rear ED, and 3) all-wheel ED, are designed and discussed based on vehicle performances during normal cornering and circling maneuvers. Through both simulation and experimental results, the three different ED can achieve almost the same vehicle performances in terms of the EGV sideslip angle, yaw rate, and trajectory. Moreover, when an ED is applied to only one pair of wheels, i.e., either the front or rear pair, the other pair of wheels can be utilized to estimate (the passive tires are adopted as sensors under normal driving conditions) the EGV longitudinal speed and yaw rate and consequently generate the reference wheel speeds for the ED wheels. Thus, sensors for measuring the vehicle speed and yaw rate to generate the reference wheel speeds in the design of ED, such as the all-wheel ED design for the EGV, may become unnecessary. Both simulation and experimental results validate the designs of the three ED.
20 March 2013 in IEEE/ASME Transactions on Mechatronics
The energy-efficient control allocation (EECA) scheme was previously proposed to distribute control efforts for overactuated systems by explicitly incorporating efficiency functions and working modes of redundant actuators. In this paper, three different real-time EECA schemes, namely adaptive EECA (A-EECA), KKT-based, and rule-based EECA, are proposed and compared for longitudinal speed tracking control of an electric ground vehicle (EGV) with two pairs of in-wheel motors. Two additional power resistor packs inserted in the dc circuits are applied to modify the operating efficiencies of two rear in-wheel motors, which are calibrated for experimental validations of the three EECA designs on a prototype EGV. In terms of the vehicle speed tracking performances, actuator dynamic responses, and total energy consumptions, both simulation and experimental results are evaluated and compared for the three distinct EECA methods. For the same EGV speed tracking effects, both simulation and experimental results indicate different power consumption savings are achieved by three EECA designs with different dynamic responses.