Search Results

The operation management of electric Taxi fleets requires cooperative optimization of Charging and Dispatching. The challenge is to make real-time decisions about which is the optimal charging station or passenger for each vehicle in the fleet. With the rapid advancement of Vehicle Internet of Things (VIOT) technologies, the aforementioned challenge can be readily addressed by leveraging big data analytics and machine learning algorithms, thereby contributing to smarter transportation systems. This study focuses on optimizing real-time decision-making for charging and dispatching in large-scale electric taxi fleets to improve their long-term benefits. To achieve this goal, a spatiotemporal decision framework using Bi-level optimization is proposed. Initially, a deep reinforcement learning-based model is built to estimate the value of charging and order dispatching under uncertainty.

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.

Automated driving system is a multi-source sensor data fusion system. However different type sensor has different operating frequencies, different field of view, different detection capabilities and different sensor data transition delay. Aiming at these problems, this paper introduces the processing mechanism of out of sequence measurement data into the multi-target detection and tracking system based on millimeter wave radar and camera. After the comparison of ablation experiments, the longitudinal and lateral tracking performance of the fusion system is improved in different distance ranges.

The Chinese government and industries have proposed strategic plans and policies for automotive renewable-energy transformation in response to China’s commitments to peak the national carbon emissions before 2030 and to achieve carbon neutrality by 2060. We thus analyze the evolution of carbon emissions from the vehicle fleet in China with our data-driven models based on these plans. Our results indicate that the vehicle life-cycle carbon emissions are appreciable, accounting for 8.9% of the national total and 11.3% of energy combustion in 2020. Commercial vehicles are the primary source of automotive carbon emissions, accounting for about 60% of the vehicle energy cycle. Among these, heavy-duty trucks are the most important, producing 38.99% of the total carbon emissions in the vehicle operation stage in 2020 and 52.18% in 2035.

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.

As the connectivity of vehicles increases rapidly, more vehicles have the capability to communicate with each other. Because Vehicular Ad-hoc NETworks (VANETs) have the characteristics of solid mobility and decentralization, traditional security strategies such as authentication, firewall, and access control are difficult to play an influential role. As a soft security method, trust management can ensure the security attributes of VANETs. However, the rapid growth of newly encountered nodes of the trust management system also increases the requirements for trust establishing mechanisms. Without a proper trust establishment mechanism, the trust value of the newly encountered nodes will deviate significantly from its actual performance, and the trust management system will suffer from newcomer attacks.

A comfortable thermal environment can alleviate fatigue, reduce irritability, and improve driving safety. However, it is rather a challenge to evaluate thermal comfort inside a vehicle due to multifarious geometric and environmental factors as well as human differences. This study conducted a series of field experiments both in summer and winter conditions, measuring the thermal environment parameters inside the compartment and the skin temperature of experimental personnel, and carrying out subjective thermal sensation and comfort questionnaires. The experimental results showed that head and trunk are the most relevant parts of all human body parts to the overall thermal sensation/comfort. For overall thermal sensation, the value of regression R2 referring to head/trunk is 0.691/0.721, while those corresponding to overall thermal comfort is 0.802/0.773.

Lane marking detection plays a crucial role in Autonomous Driving Systems or Advanced Driving Assistance System. Vision based lane marking detection technology has been well discussed and put into practical application. LiDAR is more stable for challenging environment compared to cameras, and with the development of LiDAR technology, price and lifetime are no longer an issue. We propose a lane marking detection algorithm based on solid-state LiDARs. First a series of data pre-processing operations were done for the solid-state LiDARs with small field of view, and the needed ground points are extracted by the RANSAC method. Then, based on the OTSU method, we propose an approach for extracting lane marking points using intensity information.

The accuracy of parking slot detection is a challenge for the safety of the Automated Valet Parking (AVP), while traditional methods of range sensor-based parking slot detection have mostly focused on the detection rate in a scenario, where the ego-vehicle must pass by the slot. This paper uses three-dimensional Light Detection And Ranging (3D LiDAR) to efficiently search parking slots around without passing by them and highlights the accuracy of detecting and tracking. For this purpose, a universal process of 3D LiDAR-based high-accuracy slot perception is proposed in this paper. First, the method Minimum Spanning Tree (MST) is applied to sort obstacles, and Separating Axis Theorem (SAT) are applied to the bounding boxes of obstacles in the bird’s-eye view, to find a free space between two adjacent obstacles. These bounding boxes are obtained by using common point cloud processing methods.

Tracking vehicle trajectories is essential for autonomous vehicles and advanced driver-assistance systems to understand traffic environment and evaluate collision risk. In order to reduce the position deviation and fluctuation of tracking on-road vehicle by millimeter-wave radar (MMWR), an interactive multi-model Kalman filter (IMM-KF) tracking algorithm including data association and track management is proposed. In general, it is difficult to model the target vehicle accurately due to lack of vehicle kinematics parameters, like wheel base, uncertainty of driving behavior and limitation of sensor’s field of view. To handle the uncertainty problem, an interacting multiple model (IMM) approach using Kalman filters is employed to estimate multitarget’s states. Then the compensation of radar ego motion is achieved, since the original measurement is under the radar polar coordinate system.

For the personification of automotive vehicle function performance under common traffic scenarios, analysis of human driver behavior is necessary. Based on China Field Operational Test (China-FOT) database of China Natural Driving Study project, this paper studies the driver's response in the common cut-in scenario. A total of 266 cut-in cases are selected by manual interception of driving recorder video. The relevant traffic environment characteristics are also extracted from video, including light conditions, road conditions, scale and lateral position of cut-in vehicle, etc. Dynamic information is decoded form CAN, such as speed, acceleration and so on. Then image processing results, such as relative speed and distance of cut-in and subject vehicles, are calculated. Statistical results based on above information show the response type and distribution of human driver: the behavior of keeping lane is 96.24%, in which the ratio of braking response is 51.13%.

Aiming at the problem that the evaluation model had proposed by researchers to evaluate the drivability of a vehicle in the process of engine start to exist poor stability and poor accuracy. In this paper, a drivability evaluation model combined with principal component analysis and support vector regression is proposed. In this evaluation model, the principal component analysis is adapted to determine the input index of evaluation model, and the drivability evaluation model is built on the basis of support vector regression. The experimental results demonstrate that the drivability evaluation model is proposed by this paper has higher accuracy and stability than the model using the BP neural network. This method can be as well extended to other evaluation models, with higher theoretical guidance and application value in practical issues.

The energy consumption depends not only on the structures of vehicles but also on their operating conditions. For vehicles with the same structure, the operating conditions will vary from driver to driver. In this paper, considering the difference of operating conditions, the concept of statistical energy consumption is proposed to reveal the statistical law of actual vehicle energy consumption. In this paper, a plug-in hybrid electric vehicle (PHEV) is taken as the research object. Based on the distribution law of three vehicle use factors, i.e. vehicle mass, daily driving distance and driving aggression, Monte Carlo method is used to simulate and calculate the statistical energy consumption and statistical comprehensive energy consumption. Then, the energy consumption values that only considered the daily driving distance is calculated.

In this paper, a potential risk assessment algorithm is proposed. The obvious risk assessment measure is defined as time to collision (TTC), whereas the potential risk measure is defined as the time before the host vehicle has to decelerate to avoid a rear-end collision assuming that the target vehicle brakes, i.e. time margin (TM). The driving behavior of the human driver in the dangerous car following scenario is studied by using the naturalistic driving data collected by video drive record (VDR), which include 78 real dangerous car following dangerous scenarios. A potential risk assessment algorithm was constructed using TM and the dangerous car following scenarios. Firstly, the braking starting time during dangerous car following is identified. Next, the TM at brake starting time of the 78 dangerous car following scenarios is analyzed. In the last, the thresholds of the potential risk levels are achieved.

Based on the durability cycle test of fuel cell stack and the characteristics of cyclic working conditions, this paper defines the characteristic current point and studies the attenuation rule of the fuel cell stack voltage over time under the characteristic current point. The results show that the voltage of the fuel cell stack appears to be linear downward under the characteristic current point. and the voltage attenuation rate of the fuel cell stack increases quadratically with the increase of the current density in addition to the open-circuit voltage point. Then the coefficient of variation is introduced in statistics as the index to characterize the voltage consistency attenuation of the fuel cell stack, and its variation rule is explored. The results show that the voltage consistency of vehicle fuel cell stack decreases seriously with the increase of running time under the condition of durable cycling.

Because of the high frequency and serious consequences of traffic accidents in the intersection area, it is of great significance to study the vehicle conflict and collision scenarios of the intersection area. Due to few actual crash accidents occurring in naturalistic driving studies data or field operational tests data, the data of traffic accident database should be also used to analyze the intersection conflict and collision. According to the China Field Operation Test (China-FOT) database and the China in Depth Accident Study (CIDAS) database, the distribution feature of the respective intersection scenario type is obtained from the data analysis. Based on the intersection scenario type, two characters of intersection conflict and collision, the environmental character and the vehicle dynamic character, are used to analyze for the integration process of intersection conflict and collision.

It is a worldwide technical difficulty to predict brake squeal uncertainty and its propagation regularity due to key parameters’ randomcity. However, as a widely used stochastic finite element method, Monte Carlo Method costs a large amount of time in calculation. It is very important to establish a fast prediction method for brake squeal uncertainty due to key parameters. In this paper, perturbation concept was applied for disc brake squeal uncertainty prediction. Firstly, a simplified, parameterized finite element model of disc brake was established for complex eigenvalues calculation. Then sensitivity analysis of real parts and frequencies of complex eigenvalues to influence parameters was carried out based on the finite element model. A series of second order perturbation polynomial formulas were fitted from the sensitivity analysis results, which were used for calculation of uncertain complex eigenvalues.

Faced with intricate traffic conditions, the single sensor has been unable to meet the safety requirements of Advanced Driver Assistance Systems (ADAS) and autonomous driving. In the field of multi-target tracking, the number of targets detected by vision sensor is sometimes less than the current tracks while the number of targets detected by millimeter wave radar is more than the current tracks. Hence, a multi-sensor information fusion algorithm is presented by utilizing advantage of both vision sensor and millimeter wave radar. The multi-sensor fusion algorithm is based on centralized fusion strategy that the fusion center takes a unified track management. At First, vision sensor and radar are used to detect the target and to measure the range and the azimuth angle of the target. Then, the detections data from vision sensor and radar is transferred to fusion center to join the multi-target tracking with the prediction of current tracks.

To promote the development of automated vehicles (AVs), large scale of field operational tests (FOTs) were carried out around the world. Applications of naturalistic driving data should base on correlative scenarios. However, most of the existing scenario typologies, aiming at advanced driving assistance system (ADAS) and extracting discontinuous fragments from driving process, are not suitable for AVs, which need to complete continuous driving tasks. In this paper, a systematic scenario-typology consisting of four layers (from top to bottom: trip, cluster, segment and process) was first proposed. A trip refers to the whole duration from starting at initial parking space to parking at final one. The basic units ‘Process’, during which the vehicle fulfils only one driving task, are classified into parking process, long-, middle- and short-time-driving-processes. A segment consists of two neighboring long-time-driving processes and a middle or/and short one between them.