Search Results

The around view monitor (AVM) system for vehicles usually suffers from the distortion of surrounding objects caused by incomplete rectification and stitching, which seriously affects the driver's judgment of the surrounding environment during the reversing process. In response to solve this problem, an augmented around view monitor (AAVM) system fusing image and depth information is proposed, which highlights the point clouds of persons or vehicles at the rear of the vehicle. First, an around view image is generated from four fisheye cameras. Then, the calibration of multi TOF cameras is conducted to improve their accuracy of depth estimation and obtain extrinsic camera positions. Next, the 2D-driven object point cloud detection method is proposed to localize and segment object point clouds like vehicles or persons.

The driver's face detection and tracking method important for Advanced Driver Assistance Systems (ADAS) and autonomous driving in various situations. The deep SORT algorithm has integrated appearance information, the motion model and the intersection-over-union (IOU) distance methods, and has been applied to face tracking, but it depends on detection information in every frame. Once the detection information lacks, the deep SORT algorithm will wait until the target detects bounding boxes appear again, even if the target didn’t disappear or shield. Hence, we propose to use a new tracker that not completely depend on the detection algorithm to cascade with the deep SORT algorithm to realize stable driver's face tracking. At first, the driver's face detection and tracking will be accomplished by the MTCNN-deep-SORT algorithm.

To help predict the injury responses of child pedestrians and occupants in traffic incidents, finite element (FE) modeling has become a common research tool. Until now, there was no whole-body FE model for 10-year-old (10 YO) children. This paper introduces the development of two 10 YO whole-body pediatric FE models (named CHARM-10) with a standing posture to represent a pedestrian and a seated posture to represent an occupant with sufficient anatomic details. The geometric data was obtained from medical images and the key dimensions were compared to literature data. Component-level sub-models were built and validated against experimental results of post mortem human subjects (PMHS). Most of these studies have been mostly published previously and briefly summarized in this paper. For the current study, focus was put on the late stage model development.

This paper presents an optimal cooperative path planning method considering driver’s driving intention for shared control to address target path conflicts during the driver-automation interaction by using the convex optimization technique based on the natural cubic spline. The optimal path criteria (e.g. the optimal curvature, the optimal heading angle) are formulated as quadratic forms using the natural cubic spline, and the initial cooperative path profiles of the cooperative path in the Frenet-based coordinate system are induced by considering the driver’s lane-changing intention recognized by the Support Vector Machine (SVM) method. Then, the optimal cooperative path could be obtained by the convex optimization techniques. The noncooperative game theory is adopted to model the driver-automation interaction in this shared control framework, where the Nash equilibrium solution is derived by the model predictive control (MPC) approach.

An autonomous emergency braking (AEB) system can detect emergency conditions using sensors (e.g., radar and camera) to automatically activate the braking actuator without driver input. However, during the hard braking phase, crash conditions for the restraint system can easily change (e.g., vehicle velocity and occupant position), causing an out-of-position (OOP) phenomenon, especially for unbelted occupants entering the airbag deployment range, which may lead to more severe injuries than in a normal position. A critical step in reducing the injury of unbelted occupants would be to design an AEB system while considering the effect of deployed airbags on the occupants. Thus far, few studies have paid attention to the compatibility between AEB and airbag systems for unbelted occupants. This study aims to provide a method that combines AEB and airbag systems to explore the potential injury reduction capabilities for unbelted occupants.

In order to provide an improved countermeasure for occupant protection, a new type of active reversible preload seatbelt (ARPS) is presented in this paper. The ARPS is capable of protecting occupants by reducing injuries during frontal collisions. ARPS retracts seatbelt webbing by activating an electric motor attached to the seatbelt retractor. FCW (Forward Collision Warning) and LDW (Lane Departure Warning) provide signals as a trigger to activate the electric motor to retract the seatbelt webbing, thus making the occupant restraint system work more effectively in a crash. It also helps reduce occupant’s forward movement during impact process via braking. Four important factors such as preload force, preload velocity and the length and timing of webbing retraction play influential roles in performance of the ARPS. This paper focuses on studying preload performance of ARPS under various test conditions to investigate effects of the aforementioned factors.

The combination of passive and active vehicle safety technologies can effectively improve vehicle safety. Most of them predict vehicle crashes using radar or video, but they can't be applied extensively currently due to the high cost. Another collision forecasting method is more economic which is based on the driver behavior and vehicle status, such as the acceleration, angular velocity of the brake pedal and so on. However, the acceleration and angular velocity of the brake pedal will change with the driver and the vehicle type. In order to study the effect of different drivers and vehicle types on the braking acceleration and angular velocity of the brake pedal, six volunteers were asked to drive five vehicles for simulating the working conditions of emergency braking, normal braking, inching braking and passing barricades under different velocities. All the tests were conducted on asphalt road, and comprehensive experimental design was used to arrange tests.

Panoramic images can provide critical information for Advanced Driving Assistance Systems (ADAS), such as parking spaces and surrounding vehicles. However, the vehicle in the bird's-eye view image is severely distorted and incomplete, and the visual information becomes very blurred in some illumination insufficient environments. If the driver cannot see the surrounding environment information, the risk of collision will increase, especially during parking. To better percept the local environment with the help of panoramic images, we use panoramic image segmentation results to construct a virtual surround view monitoring system to provide drivers with clearer perception information. Firstly, a lightweight segmentation network is redesigned based on SegNet, which will improve the accuracy of the segmentation without increasing the model’s inference time. Secondly, we build an augment visualization around view monitor (AV-AVM) system with regards to the segmentation results.

Due to the variation of compartment design and occupant’s posture in self-driving cars, there is a new and major challenge for occupant protection. In particular, the studies on occupant restraint systems used in the self-driving car have been significantly delayed compared to the development of the autonomous technologies. In this paper, a numerical study was conducted to investigate the effectiveness of three typical restraint systems on the driver protection in three different scenarios.

The driving risk field model offers a feasible approach for assessing driving risks and planning safe trajectory in complex traffic scenarios. However, the conventional risk field fails to account for the vehicle size and acceleration, results in the same trajectories are generated when facing different vehicle types and unable to make safe decisions in emergency situations. Therefore, this paper firstly introduces the acceleration and vehicle size of surrounding vehicles for improving the driving risk model. Then, an integrated decision-making and planning model is proposed based on the combination of the novelty risk field and model predictive control (MPC), in which driving risk and vehicle dynamics constraints are taken into consideration. Finally, the multiple driving scenarios are designed and analyzed for validate the proposed model.