Research On Vehilce Driving Safety Warning Model For Driver Assistant System

At present, it is of common occurrence that traffic accidents occurred because of driver’s misoperation and many researchers are paying remarkable attention to the driver assistant system(DAS) for improving the situation. The driver assistant system is to give driver a warning before an emergency situation in order to timely correct the improper operation and avoid or reduce the traffic accidents because of the driver’s individual factor. Thus, the research of driver assistant system has important theoretical significance and practical application value for improving safety driving and road traffic safety.On the basis of introduction about the main part of driver assistant system, as well as its review and analysis of development and research, research on vehilce driving safety warning model for driver assistant system is proposed for solving the false alarm problem of the existing driver assistant system since neglecting the driver’s individual diversity and vehicle characteristic. Through introducing the driver’s individual diversity and vehicle characteristic in this paper,the vehilce driving safety warning model for vehicle following, lane chaing, overtaking and lane departure, can accord well with the driving habits even more, so as to effectively improve warning accuracy and practicability of driver assistant system. More importantly, the paper can complement and perfect relating technology and theory of the existing driver assistant system based on the above research. Based on the study of driving safety warning for single vehicle, the driving safety warning model for a platoon of adaptive cruise control(ACC) vehicles is deep into research. Meanwhile, the dynamics, in particular the string stability, of a platoon of ACC vehicles has been studied with considering the actuator lag and sensor delay. The main research work and research results in this dissertation are summarized as follows:1. The safety warning model of car following considering lateral separation and overtaking stimulation is established. From qualitative observations of any road traffic, however, it is not difficult to notice that not every vehicle is positioned in the centre of the lane. The centerline separation between the consecutive vehicles always exists and results in staggered car following behavior. Under the circumstances, the follower decides to pass the leader by using the effective route width because of incitement degree of enough lateral clearance. In order to depict the real car-following behaviors better and ensure following or passing safety of the follower, it is quite necessary to present a new car-following safety warning model by introducing the minimum safe distance and proposing the concepts of maximum escape speed(MES) after taking into account the leading vehicle position in the road and the impact of the centerline separation between the leading vehicle and the following vehicle on the car-following behavior. Finally, the numerical simulation analysis was made using the built model by the Matlab simulation platform. By computing the minimum safe distance under different lateral offset of the two consecutive vehicles, the new car-following safety warning model can reduce or avoid traffic accidents occurred owing to blind and drastic following or passing the leader as well as extending the car following model into two-dimensional model.2. The lane changing safety warning model considering different driver style. The traditional models of lane-changing minimum safety distance lack of security and flexibility since they did not consider driver’s individual difference. In this paper, after fully analyzing the movement state of the related vehicles when a vehicle is changing lanes, all kinds of collisions between lane-changing vehicle and related vehicles are given. On this basis, in allusion to the processing characteristics of different drivers in kinds of lane-changing situation and by applying vehicle kinematics theory, a new model of lane-changing minimum safety distance based on different speeds and different types of driver, which is called ellipse model of minimum safety distance is put forward. The model can solve the above mentioned defects which exist in traditional models of lane-changing minimum safety distance. Finally, simulation was carried out on the Matlab simulation platform based on the ellipse model of lane-changing minimum safety distance. It is found that the presented safety warning model is consistent with actual situation and meets different drivers’ need of warning accuracy since that it considers the driver’s factor.3. The overtaking safety warning model considering different vehicle and driver characteristics is proposed. The time and distance required for overtaking vehicle based on the existing model can not be adapted to all vehicles and drivers since neglecting the vehicles and drivers individual characteristics. Based on this, there are two common cases occurred: rear-end collision occurred because of less time and distance required; it is not easy to overtake for overtaking vehicle because of larger time and distance required. In addition, overtaking timing and minimum safety distance between the overtaking vehicle and overtaken vehicle are distinct for different driver. Based on the existing model, in allusion to kinds of overtaking situation of different vehicle and driver types, different speed limit for different lanes, and combining vehicle dynamics performance, a new mathematic model of calculating the time required for overtaking and the overtaking distance for overtaking vehicle form beginning to the end for two-lane highway is put forward in this paper. At last, the simulation and numerical results are given and analyzed. In short, the overtaking warning model meets the need of warning accuracy and adaptivity fully since taking two influenced factors into account and is also a powerful supplement of driver assistant system.4. The lane departure warning(LDW) model considering driver’s behavior characteristics is analyzed. First, three different existing estimation models of vehicle time to lane crossing(TLC) are established, and their advantages and disadvantages, robustness and adaptability are analysed and ummarized. Then, lane departure characteristics that can meet driver’s behavior are given based on the statistics and analysis of a large sample of data form driving test. After that, the different thresholds of time to lane cross(TLC) using fuzzy control method for driver with different driving behaviors in different lanes and different vehicle movements are set. On this basis, a LDW strategy determining driver’s intentions was presented by introducing identifying vehicle movements. Finally, a vast quantify of the simulated experiments are given to demonstrate the effectiveness of the proposed LDW algorithm. The results show that LDW model has high warning accuracy, real-time and satisfaction.5. The safety and stability of a platoon of ACC vehicles has been studied. First, The impact of actuator lags and communication delay on ACC vehicles stability is analyzed based on the ideal ACC model with zero delays. Get a conclusion: both communication delay and actuator lag deteriorate the dynamic stability of vehicles and platoons but have no direct impact on static performance. At the same amplitude, the negative effect of commiunication delay is larger than that of actuator lag. Then, a new ACC model is proposed considering two delaysand the region of string stability of a platoon of ACC vehicles is found under the different actuator lag and time delay plane by analyzing the transfer function of spacing error in frequency domains. According to the stability conditions(h-τ-ζ), the practical stability of both homogeneous and heterogeneous platoons of ACC vehicles is investigated. The results show that such strategies have great potential in enhancing the stability and safety of ACC vehicles.