In this paper, we present an optimal path planning and speed control strategy for motion planning of unmanned ground vehicles to follow the desired path in various driving environments. The novelty of this work is that it integrates the longitudinal speed and lateral motion control by maximizing the longitudinal progression and minimizing the lateral path tracking errors and formulates a standard quadratic optimization problem. Firstly, a dynamic programming approach is introduced to convexify the corridor constraints in path pre-selection and replanned for obstacle avoidance. The idea is to find a convex feasible set for the optimization problem using the convex constraints. And then a standard quadratic convex problem is formulated to optimize the longitudinal motion and steering angle control in prediction and control horizon. The longitudinal progression and lateral tracking errors are decomposed in orthogonal curvilinear coordinates. The centerline as well as road borders are also described by using cubic spline curves. Some challenge scenarios including static and moving participating obstacle vehicles are tested and validated for the proposed motion planning strategy. The feasibility and performance of the proposed strategy are demonstrated by numerical experiments in terms of the stability and maneuverability of unmanned ground vehicles in different driving conditions.

This paper proposes an improved virtual reference model for semi-active suspension to coordinate the vehicle ride comfort and handling stability. The reference model combines the virtues of sky-hook with ground-hook control logic, and the hybrid coefficient is tuned according to the longitudinal and lateral acceleration so as to improve the vehicle stability especially in high-speed condition. Suspension state observer based on unscented Kalman filter is designed. A sliding mode controller (SMC) is developed to track the states of the reference model. The stability of the SMC strategy is proven by means of Lyapunov function taking into account the nonlinear damper characteristics and sprung mass variation of the vehicle. Finally, the performance of the controller is demonstrated under three typical working conditions: the random road excitation, speed bump road and sharp acceleration and braking. The simulation results indicated that, compared with the traditional passive suspension, the proposed control algorithm can offer a better coordination between vehicle ride comfort and handling stability. This approach provides a viable alternative to costlier active suspension control systems for commercial vehicles.

In this paper, a static state observer algorithm based on the static equilibrium position is proposed, which can realize accurate control of electric vehicle height adjustment with existing road excitation. The existence of road excitation can lead to deflection variation of the electronically controlled air suspension (ECAS). The use of only dynamic deflection as the reference for the electric vehicle height adjustment will produce great errors. Therefore, this paper provides an observation algorithm, which can realize the accurate control of vehicle height. Firstly, the static equilibrium position equation of suspension is derived according to the theory of hydrodynamics and characteristics of pneumatic chamber. Secondly, a vehicle dynamics model with seven degrees of freedom (7-DOF) is established and the kinetic equations are discretized. Then, the unscented Kalman filter (UKF) algorithm is used to obtain the static equilibrium position of vehicle. According to the vehicle static equilibrium position obtained by UKF, the height of the vehicle is adjusted by using a fuzzy controller. The simulation and experimental results show that this proposed algorithm can realize the control of vehicle height with an accuracy of over 96%, which ensures the excellent driving performance of vehicles under different road conditions.

The new type of transportation based on intelligent driverless vehicles will bring great changes to people’s travel modes and put forward higher requirements for the ride comfortability of vehicles. This paper presents a new observing algorithm to estimate the suspension states in real time and cooperate with sliding mode controller to improve the ride comfortability. First, the nonlinear model of an air suspension system equipped with a continuously controllable damper is described in detail. Then, this nonlinear suspension model is linearized precisely based on the differential geometry theory; a linear Kalman filter observer is implemented for this linearized model; through the coordinate reverse transformation, the designed linear observer is transformed into a nonlinear one, which will be suitable for the original nonlinear system, so that, the proposed state observer can estimate all the states of the nonlinear quarter car suspension system. Then, in this nonlinear suspension system, a model reference sliding mode controller is designed to continuously control the damping force to improve the ride comfortability. Finally, the effectiveness and advantage of the proposed feedback linearization Kalman observer is illustrated by comparing with a traditional extended Kalman filter observer. The simulation research shows that the proposed feedback linearization observer enjoys a better estimation accuracy, higher operation efficiency, and greater control performance while cooperating with the sliding mode controller in ride comfortability control.