This paper addresses the stability of traffic flow on a highway when the vehicles operate under an adaptive cruise control (ACC) system. ACC systems are commonly designed to maintain a constant time-gap between vehicles during vehicle following. Previous researchers in literature have produced contradictory results on whether the traffic flow is stable when the constant time gap spacing policy is used. This paper resolves the contradiction and shows that the boundary conditions used at the inlets and exits influence traffic flow stability in the case of the constant time-gap policy. Further, the paper shows that it is possible to design an unconditionally stable spacing policy, i.e. a spacing policy which guarantees traffic stability under all boundary conditions. The practical implications of instability are shown through traffic simulation results. The advantages of an unconditionally stable spacing policy over the constant time-gap policy are demonstrated. The answer to the question “Should ACC systems be designed to maintain a constant time gap between vehicles?” is NO. It is quite easy to develop alternate spacing policies with superior stability properties.

In this paper, a multi-objective energy management strategy with an adaptive equivalent factor is proposed to improve the fuel economy, system durability, and charge-sustenance performance of fuel cell hybrid electric vehicles. Firstly, the total hydrogen consumption and degradation cost of power sources can be calculated by flexible empirical models. Then, the multi-objective optimization problem can be transformed into an objective function, which can be solved by quadratic programming to improve the real-time performance. Furthermore, an adaptive Unscented Kalman filter is designed to estimate the aging state of the fuel cell system. The equivalent factor in the objective function can be adaptively updated by the estimated aging state, which can balance the conflict between the fuel economy and the system durability while keeping the state-of-charge in an ideal range. Finally, simulation results show that when the fuel cell system is obviously damaged during the operation, the proposed energy management strategy still can minimize the total cost and maintain the charge-sustenance performance under different driving cycles compared with other methods.

Advanced Driving Assistance System (ADAS) provides warning and automatic control intervention to a human driver for accident avoidance and collision mitigation. To simultaneously minimize the human-machine conflict and enhance human trust in the driving assistance system, ADAS should take personalized driving characteristics into account to support a particular driver in an optimal manner. Typical Personalized Driving Assistance System (PDAS) trains a driver model offline via complicated machine learning algorithms and assumes that the trained driver model maintains reliable until the next model retraining phase. However, even for a specific driver, its driving style can vary substantially due to various factors, e.g., daily mood variation, weather condition, and hardware adaption. To account for the time-varying feature of the driving style, the physics-based driver-model parameter identification has been applied to achieve an authentic PDAS. The commonly employed online driver model parameter identification approach is the Least Square (LS) method. However, LS, leaning on the classical Persistence of Excitation (PE) condition, may have slow estimation convergence. Alternatively, this paper proposes a purely algebraic method for driver model parameter identification. This new approach can achieve almost instantaneous multiple parameter identification. Moreover, by periodically resetting the proposed algebraic identifier, time-varying parameters of the driver model can also be captured online. Simulations demonstrate the excellent performance of the proposed algorithm.

For automated driving applications, exact motion planning is pivotal to obstacle-avoidance and local driving, particularly in sophisticated surroundings. Treating this problem from a differential geometric viewpoint aids in simplifying the control and planning processes. Different from the classic differential geometric methods that result in piece-wise solutions with both forward and backward moving, this paper explores the application of sinusoids in exact motion planning regarding both the positive/continuous speed requirements and the system non-holonomic constraint. By converting the vehicle kinematic model to a similar chained form, we apply sinusoidal inputs with biases to generate the reference trajectory and discuss the principles of parameter selection for a given lateral displacement. To overcome the extra constraint brought by l’Hospital rule, we introduce a pair of entry/leave transition periods to ensures an adjustable speed profile and design flexibility. The resulting trajectory reflects an strong applicability to lane-changing or evasive steering scenarios. Since the proposed method does not account for the small-time controllability because of the positive/continuous speed constraint, the reachability problem is further investigated. We provide an analytic estimation of the reachable area edge and compare it with the numerical simulation results. Finally, some indoor field-tests verified the feasibility of the proposed method.

In this paper, we present a continuously tunable stiffness arm for safe physical human-robot interactions. Compliant joints and compliant links are two typical solutions to address safety issues for physical human-robot interaction via introducing mechanical compliance to robotic systems. While extensive studies explore variable stiffness joints/actuators, variable stiffness links for safe physical human-robot interactions are much less studied. This paper details the design and modeling of a compliant robotic arm whose stiffness can be continuously tuned via cable-driven mechanisms actuated by a single servo motor. Specifically, a 3D printed compliant robotic arm is prototyped and tested by static experiments, and an analytical model of the variable stiffness arm is derived and validated by testing. The results show that the lateral stiffness of the robot arm can achieve a variety of 221.26% given a morphing angle of 90°. The study demonstrates that the compliant link design could be a promising approach to address safety concerns for safe physical human-robot interactions.

Lidar sensors are fundamental and crucial devices for self-driving vehicles, due to their capabilities of environmental information acquisition to develop feasible driving areas and motion decisions. Even though lidars provide 3D point clouds for accurate positions, the complex data processing takes arduous efforts for online calculation, including segmentation, filtering, clustering and verification. Furthermore, sensor fusion with multiple lidar sensors (such as Velodyne HDL-32c and Ibeo-Lux 4L) aggravates the computational load, which prevents exact surrounding detection from being applied for planning and control in real time. To address the problem, this paper proposes a systematic filtering algorithm based on occupancy rates of two categories obstacle vehicle detection method for autonomous ground vehicles, considering lidar calibration, ground segmentation, ego-vehicle removal of 3D lidar point clouds. Another adaptive searching (AS) algorithm on density-based spatial clustering of applications with noise (DBSCAN) is proposed to coordinate the characteristics of scanned points with respect to distances to the lidar set-up point. An indoor perception test with a fully-instrumented autonomous hybrid electric vehicle within complicated surroundings was conducted, with 3D point cloud fused data provided by one Velodyne HDL-32c lidar sensor and three Ibeo-Lux 4L scanners, which has verified the effectiveness of the proposed approach for obstacle detection.

Tire inflation pressure affects both tire longitudinal and lateral stiffness and thus may impose a considerable influence on vehicle dynamics and handling performance. This paper presents a comprehensive study revealing the effects of tire pressure variations and their distribution among four tires on vehicle dynamics and handling performance. An extended Magic Formula tire model and a modified UniTire model involving tire inflation pressure are employed to describe the tire longitudinal and lateral forces, respectively. Two groups of vehicle maneuvers are simulated in CarSim: a single lane change maneuver with braking and a double lane change maneuver, to exhibit the effects of tire inflation pressure. Various tire pressure variations including all four tires at same and different pressures are examined. A vehicle dynamics, lateral motion stability index, and driver steering workload are utilized to quantify the influence of tire pressure variations and distributions. Analyses on the simulation results indicate that: 1) a front tire pressure reduction induces vehicle understeering tendency and a larger steering angle; 2) a rear tire pressure reduction causes oversteering characteristics and a sacrifice on vehicle stability with a larger vehicle sideslip angle; 3) all-tire inflation pressure decrease will increase driver’s steering workload; and 4) lower rear-tire inflation pressure can promote the combined performance of vehicle path-tracking and driver’s steering workload.

Non-smooth structured robust controller design has drawn a lot of attention recently due to its ability to deal with uncertainty and its convenience for implementation. In this paper, the method is extended to design the structured robust linear parameter-varying (LPV) estimator by pulling out scheduling variables from estimator using linear fractional transformation (LFT). The structured robust LPV estimator is then applied to vehicle sideslip angle estimation. Both the measured vehicle speed and estimated tire cornering stiffness are treated as scheduling variables to further reduce sideslip angle estimation error. The effects of estimator order and number of repetitiveness of scheduling variables are studied using a MATLAB/Simulink bicycle model. The developed approach is later verified in Hardware-in-the-Loop (HIL) simulation environment using dSPACE SCALEXIO and MicroAutoBox. A comprehensive high-fidelity dSPACE automotive simulation models (ASM) vehicle model is used for the real-time HIL simulation. Double-lane change and sine steer maneuvers have been implemented to verify the effectiveness of the structured robust LPV sideslip angle estimation method.

Unlike an autonomous driving system, a shared control system keeps the driver in the loop to exploit human’s decision-making, path-planning, and control skills in vehicle motion control. In the meantime, a shared control system consistently provides support to the driver in hazardous situations. Typical shared control systems concentrate on altering vehicle steering actions through either active front steering or haptic steering torque. However, an active front steering system requires costly hardware and a haptic steering torque explicitly interferences with the driver hand-wheel steering. To overcome the drawbacks of the steering-based strategies, this paper introduces a driver support system based on direct yaw moment control for highspeed collision avoidance maneuvers. The new system maintains a hierarchical structure with a high-level lateral velocity planner and a low-level lateral velocity governor. In addition, the personally-preferred collision avoidance trajectories of each driver are taken into account to make the controller personalizable. Carsim-Simulink joint simulations demonstrate that this novel shared control system can successfully assist the driver in high-speed collision avoidance maneuvers. Moreover, both physical and mental loads of the driver are substantially reduced in contrast to using an active front steering system.

In this paper, a control design for a flexible link co-robot with safety constraints is proposed. The safety constraints are converted to the constraints on the tip position and velocity. To handle this constrained control problem, a barrier Lyapunov function (BLF) is employed in the control design. The derivative of the logarithmic BLF is more complicated compared with the derivative of a quadratic Lyapunov function, which makes the problem of “explosion of terms” more serious. Thus, the dynamic surface control is used to deal with the problem. Furthermore, an extended state observer is adopted to estimate and compensate the uncertainty and disturbance in the system. The stability analysis via the singular perturbation theory shows the local practical exponential stability of the system. Simulation results indicate that the control performance is guaranteed without violation of the constraints.

There exist two intrinsic shortcomings on model predictive control (MPC) strategy, namely the extensive online calculation burden and the complex tuning process, which prevent MPC from being applied to a wider extent. To tackle these two drawbacks, different methods were proposed with majority of them treating these two issues independently. However, parameter tuning in fact has double-sided effects on both controller performance as well as real-time computational burden. Due to the lack of theoretical tools for globally analyzing the complex conflicts among MPC parameter tuning, controller performance optimization as well as computational burden easement, a look-up table based online parameter selection method is proposed in this paper to help a vehicle track its reference path under both the stability and computational capacity constraints. Matlab-CarSim conjoint simulation shows the effectiveness of the proposed strategy.