Vehicle Tracking System for Intelligent and Connected Vehicle Based on Radarand V2V Fusion

On Road Vehicle Breakdown Assistance Finder Project

Vehicle Tracking System for Intelligent and Connected Vehicle Based on Radarand V2V Fusion
The environment perception plays a significantly role in intelligent vehicles and the advanced driver assistance system (ADAS), which enhances the driving safety and convenience. Target tracking is one of the key technologies ofenvironment perception. The on-board sensors such as cameras and radar are commonly used for target tracking while they have limitations in terms of detection range and angle of view. One way to overcome the perception limitations of on-board ranging sensors by incorporating the vehicle-to-vehicle (V2V) communication.This paper proposes a vehicle tracking system which fuse the radar and V2V information to improve the target tracking accuracy. The proposed systemintegrates the radar, GPS and DSRC communication equipment. The GPS and radar are utilized to obtainits own position information and the position information of nearby vehicles.The proposed system also resolves the problem of data association in multiple target tracking measurements by other connected vehicles' identity information. With the association measurements, a Kalman filter is used to improve the accuracy of target tracking. Anassessment of tracking system in real road environment shows that the proposed fusion approach for target tracking can reduce the data association error and improve the vehicle target tracking accuracy.  
Driving safety has been a crucial element in the design of future intelligent transportation systems (ITS). The increasing popularity of intelligent vehicle (IV) and the advanced driver assistance system (ADAS) offers the potential to significantly enhance driving safer, more efficient and more comfortable. Target tracking is one of the key technologies in intelligent vehicle environment perception, which can prevent collisions and save lives. The different on-board sensors such as cameras [1,2], laser lidar [3] and radars [4] are equipped to track nearby targets and improve the perception of the vehicle at its surroundings. However, the main limitation of these on-board sensors is the accuracy of the measurements which highly depends on the type and quality of the sensor. For example, the performance of sensors is affected by the field of view, detection range and on-board vehicle platforms.

https://codeshoppy.com/shop/product/on-road-vehicle-breakdown-assistance-app/

With the development of the vehicle-to-vehicle (V2V) communication technology, various sensors are embedded on vehicles. These sensors can collect data from their surrounding environment, and then, transmit relevant information to connected equipment using vehicular communication technologies. In the paper, these vehicles which can exchange the information via dedicated short-range communications (DSRC) are called connected  vehicle. In 2006, A plan called the vehicle safety communications applications (VSC-A) [5], which sponsored by the United States department of transportation (USDOT), is used to overcome the limitation of environment perception for intelligent vehicles. However, the diversity and huge data from on-board sensors and connected vehicle will cause the data uncertainties problems in multi-sensor fusion [6]. The uncertainty of the target tracking data is mainly manifested in the data association. Data provided by sensors are always affected by some clutter information in the measurement, the number of measurement targets will be more than the number of actual targets. Therefore, the main challenge of multi-target tracking is to associate multiple measurements with its historical trajectory. The common measurement-measurement association algorithms are nearest neighbor (NN), multiple hypothesis racking (MHT) [7] and joint probabilistic data association (JPDA) [8]. JPDA algorithm is one of the most well-known multi-target tracking algorithms. However, with the increase of target number and candidate target number, the feasibility calculation of JPDA will be exponential growth. Related Work. Radar has been widely used in detecting vehicle in many ADAS [9]. The radar frequency shift in the received signal is used to measure the distance to the detected object. Detected objects are then tracked and filtered based on motion characteristics to identify vehicles and other obstacles [10]. Literature [10] pointed out that the measurements of radar are quite noisy, requiring extensive filtering and cleaning. What’s more, the radar tracking only depends on the relative motion. In recent years, the fusion
method of radar and V2V communication has been shown more effective in vehicle detection and tracking. In [11], the PHD filter algorithm was used for multiple vehicle cooperative localization. In [12], the authors proposed a cooperative positioning fusion method based on on-board radar and V2V communication. In [13], a position algorithm which can reduce the uncertainty of the position estimate of the host vehicle is proposed. A review on different cooperative and non-cooperative positioning sensor for intelligent vehicle was presented in [14]. However, the authors in [11-14] did not consider multiple measurements in target tracking, and how to utilize CVs' identity information to improve the data association accuracy. This paper proposes a vehicle tracking system that fuses radar and V2V information to improve the vehicle tracking accuracy. The proposed system fuses the vehicle targets obtained by radar and the GPS and identity information received via the DSRC transceiver from other CVs. The problem of data association in multiple target tracking measurements is solved by CVs' identity information. With the association measurements, a Kalman filter is used to improve the accuracy of target tracking. Code Shoppy