Implementation of Advanced Driver Assistance System (adas) to Eliminate The Driver’s Negligence

According to the World Health Organization more than 1.25 million people die each year as a result of road traffic crashes. Injuries caused by road accidents result in considerable economic losses to individuals, their families, and to nations as a whole. For most of the cases, these accidents are caused due to driver negligence. In order to decrease this rate of accidents, researchers in the field of automotive industry are keen to resolve this issue. One of the most innovative approaches is Advanced Driver Assistance System (ADAS), which is trying to eliminate the driver’s negligence by introducing different mechanism. This can not only reduce the rate of accidents but also ensure the safety of passengers. Emergency braking and blind spot detection are two of many other mechanisms which can be helpful for achieving this objective. Collision avoidance system also proves to be a valuable asset for automotive industry since it can take timely action if the driver is not responsive.

ADAS relies on radars and camera sensors. Radar has proven to be useful against different weather conditions. Most of the automotive radars are based on Frequency Modulated Continuous Wave (FMCW). The main reason behind this is that FMCW radar can calculate range and velocity of multiple targets simultaneously with good resolution. For autonomous driving high resolution radar sensors are key components, but having the drawback of high data rates. In order to reduce the amount of sampled data, random samples can be omitted. For estimating the missing data, several compressed sensing reconstruction techniques are used in order to recover the information. The problem with these techniques is that they require large amount of iterations and thus cannot be useful for the real world scenarios. The goal of the thesis is to resolve this issue and to evaluate these compressed sensing techniques for automotive radar and analyze the influence of different parameters on the reconstruction result. Furthermore, the focus is to reconstruct the signal with minimum cost. This can be achieved with the help of comparison among different reconstruction algorithms based on the quality measures. Apart from compressed sensing, problem of interference between the signals can also occur and cause degradation of the signal to noise ratio on the receiving end and hence, set severe limitations on the radar’s detection capabilities.

Due to this, the probability of detecting weak targets reduces because of missing information. The red car has a radar mounted on the bumper. It receives an echo of the green car, but at the same time it also receives a signal from the yellow car. This will create disturbance in the received signal of the red car and hence results in interference and missing data. There are different interference cancellation techniques to overcome this problem. But, it has disadvantage as well. While canceling out the interference, it could not recover the information from that interfered part. So, these missing data need to be recovered with help of reconstruction algorithms.