基于行驶车速的车辆防撞时间预警算法

Anti-collision warning time algorithm based on driving speed of vehicle

针对一般车辆碰撞时间(time to collision,TTC)算法预警阈值固定造成车辆低速行驶中出现预警过早及高速行驶出现预警不及时的问题,该文提出一种基于行驶车速的车辆防撞时间预警方法。行驶车辆通过车载设备实时获取自车与他车的状态信息,根据车辆状态信息建立高斯平面坐标系获取车辆位置坐标,对车辆可能发生的碰撞进行分类处理,依据车辆行驶速度设定相应的安全防撞时间,然后将车辆发生碰撞需要的时间与安全防撞时间进行比较,存在碰撞危险则通过预警显示提醒驾驶员。试验结果表明:该方法预警准确率达88.89%,而一般TTC固定阈值方法则预警过早率达81.48%,预警过晚率达70.37%,故该方法对进行车辆危险预警更有效,更符合实际车辆防撞情形,提高了车辆行驶的安全性,可为车辆的及时预警提供参考。

In order to solve the problem that the warning threshold fixation of time-to-collision （TTC） algorithm causes prematurely warning at low speed and warning with a delay at high speed, this paper presents a safety time warning method of anti-collision based on vehicle driving speed. Firstly, vehicles can get the driving status information of themselves and others timely by on-board equipment. On-board equipment can get the information about latitude, longitude, speed and heading angle of the vehicle in real time. Then, according to the driving status information, vehicle coordinates information can be obtained by Gauss plane rectangular coordinate system established. After that, the possible collisions can be classified and processed. In reality, all vehicles are not required to carry out vehicle collision warning, so it can reduce vehicle collision processing and optimize the performance of the algorithm by classifying the possible collisions of vehicles, and it also can facilitate the use of different collision warning algorithms for different types. Mathematical model can be used to classify the possible collisions. Collision is divided into linear collision and side impact, and linear collision includes frontal and rear collision. For linear collision, the precondition of collision is that the lateral distance of the 2 vehicles is less than the width of vehicle. For the side impact, vehicles maintain current speed and direction, and according to the current state of vehicles, it can predict future vehicle running track, determine vehicle collision point and calculate vehicle collision time difference, and the collision time difference is used to determine the initial conditions of side impact. Next, the initial condition of vehicle collision is met, vehicle anti-collision safety time can be set based on vehicle driving speed, and then it will be compared with the time required for vehicle collision occurring for each vehicle. If the risk of collision exists, the application equipment displays warnings to remind drivers. The safety anti-collision time is the shortest time for drivers to take measures to avoid the danger. Response time of the driver is the time when the alarm signal is received, and response time of the braking system includes the reaction time of the braking system, the braking coordination time and the continuous braking time. Based on this, it is possible to get early warning of collision risk from the driver＇s reaction to braking reaction. Through the different vehicle speed, it can determine the different safety collision time, and through detecting vehicle speed changes, it can timely change vehicle anti-collision safety time threshold. Entire system can receive the vehicle status information through the on-board equipment in real time and continuously. When the precondition of linear collision and side collision is satisfied, the time needed for vehicle collision is calculated in real time, and then compared with the time threshold of safety warning. If there is a risk of collision, it will warn on the application device. Finally, experiments are carried out to verify the performance of the proposed method. In the simulation experiment, compared with the general TTC fixed threshold method, this method reduces the false alarm rate by up to 9.04%, and decreases the missing alarm rate by 8.3%. In real vehicle test, experimental results show that the suitable early warning rate of this proposed method is 88.89%, and premature warning rate of general TTC fixed threshold method reached 81.48%, and late warning rate reached 70.37%. It can be concluded that the proposed method in this paper is more effective for vehicle hazard warning, more in line with the actual situation of the vehicle anti-collision, and improves the safety of vehicle driving.