Search Results

Environmental perception is a crucial subsystem in autonomous vehicles. In order to build safe and efficient traffic transportation, several researches have been proposed to build accurate, robust and real-time perception systems. Camera and LiDAR are widely equipped on autonomous self-driving cars and developed with many algorithms in recent years. The fusion system of camera and LiDAR provides state-of the-art methods for environmental perception due to the defects of single vehicular sensor. Extrinsic parameter calibration is able to align the coordinate systems of sensors and has been drawing enormous attention. However, differ from spatial alignment of two sensors’ data, joint calibration of multi-sensors (more than two sensors) should balance the degree of alignment between each two sensors.

Recently, electro-hydraulic brake systems (EHB) has been developed to take place of the vacuum booster, having the advantage of faster pressure build-up and continuous pressure regulation. In contrast to the vacuum booster, the pressure estimation for EHB is worth to be studied due to its abundant resource (i.e. electric motor) and cost-effective benefit. This work improves an interconnected pressure estimation algorithm (IPEA) based on actuator characteristics by introducing the vehicle dynamics and validates it via vehicle tests. Considering the previous IPEA as the prior pressure estimation, the wheel speed feedback is used for modification via a proportional-integral (PI) observer. Superior to the IPEA based on actuator characteristics, the proposed PEA improves the accuracy by more than 20% under the mismatch of pressure-position relation.

Torque Vectoring Control for distributed drive electric vehicle is studied. A handling improvement algorithm for normal cornering maneuvers is proposed based on state variable feedback control: Yaw rate feedback together with steer angle feedforward is employed to improve transient response and steady gain of the yaw rate, respectively. According to the feedback coefficient's influence on the transient response, an optimization function is proposed to obtain optimum feedback coefficients under different speeds. After maximum feedforward coefficients under different speeds are obtained from the constraint of the motor exterior characteristic, final feedforward coefficients are calculated according to an optimal steering characteristic. A torque distribution algorithm is presented to help the driver to speed up during the direct yaw moment control.

A dynamic controller is designed for unmanned skid-steering vehicle. The vehicle speed is controlled through driving torque of engine to achieve the desired vehicle speed and the steering is controlled through hydraulic braking on each side of the vehicle to achieve the desired yaw rate. Contrary to the common approaches by considering non-holonomic constraints, tire slip and saturation of actuators torque influencing the driving and braking are considered, based on the analysis of vehicle dynamic model and nonlinear tire model. Hence, with conditional integrators, the dynamic controller overcoming integral saturation is designed to ensure the accurate tracking for desired signals under influence of tire forces and constraint of actuators. In addition, the exponential kind filter is utilized to enhance the ability of smoothing noise of wheel speed. To perform small radius cornering maneuvers, a dynamic control strategy for steering when vehicle speed is zero is also designed.

In recent years, intelligent driving technology is being extensively studied. This paper proposes a path planning method for perpendicular parking based on vehicle kinematics model using MPC optimization, which aims to solve the perpendicular parking task. Firstly, in the case of any initial position and orientation of the vehicle, judging whether the vehicle can be parked at one step according to the location of the parking place and the width of the lane, and then calculating the starting position for parking, and use the Bezier curve to connect the initial position and the starting position. Secondly, reference parking path is calculated according to the collision constraints of the parking space. Finally, because the parking path based on the vehicle kinematics model is composed of circle arcs and straight lines, the curvature of the path is discontinuous. The reference parking path is optimized using Model Predictive Control (MPC).

A tractor-trailer vehicle (TTV) consists of an actuated tractor attached with several full trailers. Because of its nonlinear and noncompleted constraints, it is a challenging task to avoid collisions for path planner. In this paper, we propose an efficient method to plan an optimal trajectory for TTV to reach the destination without any collision. To deal with the complicated constraints, the trajectory planning problem is formulated as an optimal control problem uniformly, which can be solved by the interior point method. A novel incremental optimization solving algorithm (IOSA) is proposed to accelerate the optimization process, which makes the number of trailers and the size of obstacles increase asynchronously. Simulation experiments are carried out in two scenarios with static obstacles. Compared with other methods, the results show that the planning method with IOSA outperforms in the efficiency.

Visual sensors are widely used in autonomous vehicles (AVs) for object detection due to the advantages of abundant information and low-cost. But the performance of visual sensors is highly affected by low light conditions when AVs driving at nighttime and in the tunnel. The low light conditions decrease the image quality and the performance of object detection, and may cause safety of the intended functionality (SOTIF) problems. Therefore, to analyze the performance limitations of visual sensors in low light conditions, a controlled light experiment on a proving ground is designed. The influences of low light conditions on the two-stage algorithm and the single-stage algorithm are compared and analyzed quantificationally by constructing an evaluation index set from three aspects of missing detection, classification, and positioning accuracy.

Recent researches in autonomous driving mainly consider the uncertainty in perception and prediction modules for safety enhancement. However, obstacles which block the field-of-view (FOV) of sensors could generate blind areas and leaves environmental uncertainty a remaining challenge for autonomous vehicles. Current solutions mainly rely on passive obstacles avoidance in path planning instead of active perception to deal with unexplored high-risky areas. In view of the problem, this paper introduces the concept of information entropy, which quantifies uncertain information in the blind area, into the motion planning module of autonomous vehicles. Based on model predictive control (MPC) scheme, the proposed algorithm can plan collision-free trajectories while actively explore unknown areas to minimize environmental uncertainty. Simulation results under various challenging scenarios demonstrate the improvement in safety and comfort with the proposed perception-aware planning scheme.

Many learning-based methods estimate ego-motion using visual sensors. However, visual sensors are prone to intense lighting variations and textureless scenarios. 4D radar, an emerging automotive sensor, complements visual sensors effectively due to its robustness in adverse weather and lighting conditions. This paper presents an end-to-end 4D radar-visual odometry (4DRVO) approach that combines sparse point cloud data from 4D radar with image information from cameras. Using the Feature Pyramid, Pose Warping, and Cost Volume (PWC) network architecture, we extract 4D radar point features and image features at multiple scales. We then employ a hierarchical iterative refinement approach to supervise the estimated pose. We propose a novel Cross-Modal Transformer (CMT) module to effectively fuse the 4D radar point modality, image modality, and 4D radar point-image connection modality at multiple scales, achieving cross-modal feature interaction and multi-modal feature fusion.

SLAM (Simultaneous Localization and Mapping) plays a key role in autonomous driving. Recently, 4D Radar has attracted widespread attention because it breaks through the limitations of 3D millimeter wave radar and can simultaneously detect the distance, velocity, horizontal azimuth and elevation azimuth of the target with high resolution. However, there are few studies on 4D Radar in SLAM. In this paper, RI-FGO, a 4D Radar-Inertial SLAM method based on Factor Graph Optimization, is proposed. The RANSAC (Random Sample Consensus) method is used to eliminate the dynamic obstacle points from a single scan, and the ego-motion velocity is estimated from the static point cloud. A 4D Radar velocity factor is constructed in GTSAM to receive the estimated velocity in a single scan as a measurement and directly integrated into the factor graph. The 4D Radar point clouds of consecutive frames are matched as the odometry factor.