Search Results

At present, the research on fatigue driving at home and abroad mainly has the following three methods: (i) driving behavioral (vehicle-based), (ii) driver behavioral (video-based), and (iii) driver physiological signals measure. The physiology-based methods have the highest recognition result. When drivers are in a state of fatigue, the Autonomic Nervous System (ANS) activity will be reflected from the physiological signal. Most of the contact sensors are used to obtain the physiological signal information of the driver. However, the contact sensors will affect the driver's driving operation, so this paper uses the frequency-modulated continuous-wave (FMCW) radar to collect the physiological signals. A fatigue driving simulation experiment was designed to collect experimental subjects' physiological signal data and separate the steady heartbeat and respiratory signals.

Lateral control is an important part in the system of driverless mining trucks, which is used to realize accurate tracking of planned path. To solve the problem of poor accuracy of the existing single point preview algorithm, firstly, the lateral error model and the simplified truck dynamics model were built. The established truck dynamics model was verified and compared by simulation. The results show that the truck dynamic model in this paper retains accurate even at higher speed. Secondly, against the time delay of truck steering system, the cascade LQR-PID controller and MPC-MRAC controller are designed. The former resists the disturbance of steering time delay through the inner PID loop, while the latter realizes the adaptive control by establishing the steering model. Then, the dual-shift condition simulation was carried out by co-simulation model, and two controllers were compared and analyzed.

Combining the advantages of GPS and INS, GPS/INS has been widely used in vehicle localization. However, there are GPS failures in actual scenes, and even a short-term failure will cause the positioning accuracy of the low-cost MEMS-SINS to decline rapidly. Therefore, aiming at the problem of vehicle positioning under short-term GPS failure, the following three aspects of work are mainly carried out. Firstly, the unscented quaternion estimator is introduced, and the velocity-combined GPS/SINS based on this algorithm is realized. Secondly, for the positioning problem under short-term GPS failure, a USQUE-LSTM positioning method is proposed by introducing the long short-term memory neural network, which is the main contribution of this paper.

In order to solve the problems of accuracy, comfort and robustness of driverless vehicles under parallel parking condition, a control method of path tracking based on model predictive control (MPC) is studied. The kinematics model of driverless vehicle under parking condition is established. The calculation method of minimum parking space size required for parking is proposed. The linear error model of vehicle kinematics is established. In order to make the vehicle track the desired path quickly and smoothly, an appropriate objective function is designed. In rolling optimization, the constraint conditions of velocity and front wheel steering angle are imposed on the objective function to achieve the solution in the control period, the control input constraint and control increment constraint are set. In order to ensure the stability of the path tracking process, constraint condition of velocity is set.

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.

Lane Centering Control System (LCCS) is a lateral Advanced Driving Assistance System (ADAS) with low acceptance. One of the main reasons is that the centering trajectory can’t satisfy different drivers, which is more obvious in the curve condition. So LCCS adaptive to different drivers needs to be designed. The trajectory planning module is an important part for LCCS. It generates trajectory according to the road information for the vehicle control module to track. This paper uses road information obtained from the scenario established in Prescan, and the trajectory planning method proposed can generate trajectories for different drivers in the curve condition. To achieve the goal, this paper proposes a trajectory planning method which contains lateral path planning and longitudinal speed planning. Firstly, the overall strategy of “road equidistant segments division” is used to describe the road information.

Vehicle driving behavior recognition is critical to improve the safety and rationality of autonomous vehicle decision-making and planning in heterogeneous vehicle mixed scenarios. Aiming at the problems that traditional driving behavior recognition methods only consider a single driving behavior, which has insufficient recognition accuracy, and the lack of consideration of the impact on the neighborhood between traffic subjects, the algorithm robustness is poor. A driving behavior recognition method based on Gaussian mixture hidden Markov model (GMM-HMM) is proposed. Firstly, preprocess the NGSIM data sample, take the surrounding vehicles lateral displacement, lateral speed are taken as the HMM observation sequence, the HMM driving behavior recognition model is established. Then, the Baum-Welch algorithm and the Viterbi algorithm are used to train the parameters of the HMM to obtain Hidden state sequence of driving behavior.

With the continuous development of intelligent transportation, the key of intelligent transportation is forward vehicle location. Accurate and reliable forward vehicle location results are of great significance to the safe driving of intelligent vehicles. In order to solve the problem of forward vehicle location and tracking, this paper proposes a method of license plate feature point detection based on convolutional neural network. Firstly, according to the prior size information of the license plate, the efficiency of license plate detection is improved by modifying the prior frame size of Tiny-YOLOv3 and improving the network structure of Tiny-YOLO-v3. Secondly, based on the structure of the residual neural network, a residual neural network was designed to detect the feature points of the license plate with a fixed aspect ratio. In this model, the anchor frame-free object detection method is applied to license plate detection.

Pedestrian passive safety and active safety both develop rapidly, such as new structural hoods/airbags for pedestrian protection and emergency automatic braking/forward collision warning are used in advanced driver assistance system (ADAS). In this study, improved pedestrian passive safety is to obtain optimal hood structural parameters and add an active pop-up hood. Headform impactor, hood model, simplified vehicle and head impaction models were established, and nine key test points were selected for crash simulation tests. After the simulation, the pedestrian protection performance of the initial hood is evaluated and analyzed based on the head injury criterion (HIC) values. Combined with the orthogonal experimental design method, this study acquired the best structural parameters scheme and applied to the active pop-up hood.

With the continuous improvement of positioning technology and industry demand, the shortcomings of each sensor are constantly amplified. Only relying on a single sensor, the demand of high-precision positioning and mapping for intelligent vehicles is difficult to be satisfied. The accuracy of system positioning and mapping is reduced due to the loss of feature points in pure visual SLAM as the environmental characteristics are not obvious or the texture is not abundant. IMU is a sensor with low cost and high update frequency, which can correct the running trajectory in real time and reduce the error of environmental factors on visual sensor data. Therefore, a method based on ORB_SLAM2 algorithm and VINS-Fusion algorithm, the stereo camera information and inertial measurement unit information are extracted and fused in robot operating system is proposed.

In order to solve the problem of automatic lane change in straight road environment, a Segment-orienting rapidly-exploring random trees algorithm (SO-RRT) is designed based on the idea of Fast-biasing RRT algorithm and piecewise design. Firstly, the local map is preprocessed, and then the control points are obtained by geometric analysis of the map according to the obstacle vehicle around the vehicle and the speed of the vehicle. On this basis, the Fast-bias RRT is carried out in segments, and the obtained path is optimized and fitted. After processing, the planned path is finally obtained, which ensures that the planned path meets the kinematic constraints of the vehicle and is as close as possible to the optimal solution. CarSim was used for vehicle and road modeling and co-simulation with Matlab / Simulink. The path planning was carried out in Simulink and the hybrid algorithm combining model predictive control algorithm and PID algorithm was used to realize path tracking.

This paper proposed a lane-change planning method for autonomous vehicle, aiming at fast obstacles avoidance in a way that make smooth and comfortable. The panning algorithm consists of dynamic programming and closed-loop forward simulation. The dynamic programming (DP) was employed to fast search a reference trajectory that avoids obstacles in topological configure space. And the closed-loop forward simulation (CFS) was used to track the reference trajectory for generating smooth trajectory, since the CFS being able to incorporate any nonlinear law and nonlinear vehicle constraints. Furthermore, an anti-windup lateral controller was designed to make the closed-loop forward simulation robust, as the controller being proved to be stable by Lyapunov function. Finally, the numerical results are provided to illustrate the effectiveness of the proposed method.

Accurate and online capacity estimation is of extreme importance to maintain the continuous operation of lithium-ion batteries. This paper proposes an indirect capacity estimation method based on the incremental capacity features and model interpretability. First, the current and voltage data of the battery are collected in real-time to construct the incremental capacity curves. The dual filter, which consists of a moving average and a Gaussian filter, is then used to smooth the curves. To achieve satisfying filtering effects, the filter window size and calculation frequency are determined by comparing different sets of conditions. 15 multiple alternative features related to the curve peak position and area are extracted. The Shapley value method based on cooperative game theory is introduced to reduce the dimension of the feature vector and to determine the key features.

In order to improve the performance of automatic emergency steering and collision avoidance of intelligent vehicle, two automatic steering control methods under ideal model following control are proposed. The two ideal reference models are the reference model with zero sideslip angle of vehicle gravity center and the reference model with no phase-lag in vehicle lateral acceleration. The control system adopts the combination of outer loop and inner loop. In the design of the outer loop controller, the optimal control is used to get the steering wheel angle needed to avoid collision. The inner loop controller uses feedforward and feedback control to get the required front and rear wheel steering angles. Taking vehicle two degrees of freedom (DOF) lateral dynamics model as the research object, the vehicle collision avoidance reference trajectory is obtained through the fifth-degree polynomial.

Abstract In lean burn combustion, jet ignition is such a promising technology that is often used to improve the ignition and enhance combustion stability when ignition failure and misfire might occur. Four different orifice configurations are applied in the fueled prechamber and lean main chamber system to investigate the effect of orifice configurations on flame propagation and pressure oscillation. The combustion phases of jet ignition, pressure wave propagation, and mechanism of jet combustion with various orifice configurations considered are presented. In the present work, three phases of jet ignition propagation are observed. Besides, it is found that when the total orifice area is the same, the single-orifice configuration presents the fastest flame propagation and pressure oscillation. The viscosity effect and interactions between jet flows in multi-orifice configurations account for this phenomenon.

This specification establishes requirements for solvent-based and water-based release agents, in bulk or aerosol form.

This specification covers a corrosion and heat-resistant alloy in the form of welding wire.

This specification covers a corrosion and heat-resistant nickel alloy in the form of welding wire.

This data dictionary provides definitions for quantities, measurement units, reference systems, measurands, measurements, and quantity modalities commonly used in the command and control of cyber-physical systems. A cyber-physical system is an engineered system that is built from, and depends upon, the seamless integration of computational algorithms and physical components. Cyber-physical systems are often interconnected via data links and networks. The term encompasses intelligent vehicles and devices that operate in any environment, including robotic and autonomous systems.