Now, it is very difficult to control the road and traffic conditions in a city. Due to increase in traffic vehicle accidents are increasing nowadays. We proposed many methods to solve this problem. But several proposed techniques needs dedicated hardware like GPS devices and accelerometers to be fixed in vehicles or fix cameras on roadside and near traffic signals. But these techniques are expensive because it needs more human effort and its monetary cost is very high. These techniques need more human effort and their monetary cost is very high, so these techniques are expensive for our society. To reduce monetary cost and human effort we are developing the proposed Wolverine method, which controls the traffic with the help of sensors in the smart phones. It is a non-intrusive method. At first we will study about how to improve the algorithm based on using GPS and magnetometer, accelerometer sensor readings for traffic and road conditions detection. We are specifically interested in identifying braking events such as frequent braking will indicate congested traffic conditions and we can characterize the type of road with the help of bumps on the roads. To check the effectiveness of the proposed method, we applied this method on the roads in Mumbai and got success.

Various methods have been proposed for activity detection in various environments, Indoor localization, traffic detection and detecting activity of a person with the help of sensors in smart phones. We can reduce the need for specialized hardware installed in vehicles or on the road side by using the smart phone based traffic estimation methods. When number of smart phone users is growing at a rapid pace, these crowd sourced solutions will get high scalability with the help of distributed participatory data collection. The GPS sensors available in user’s smart phones, GSM Radio, accelerometer and GPS sensors are used in the Nericell system. The orientation of the phone could be arbitrary with respect to the direction of motion in a smart phone based method and can be changed repeatedly.

Hence, it is required to virtually reorient the vehicle with respect to the axes of the phone. For this Nericell system uses accelerometer and GPS readings. The direction of gravity is used to sense the vertical orientation and horizontal orientation is calculated with the help of the acceleration recorded during a braking event. A system which is used to track the stolen property and works same as Nericell system is called Auto witness system. Then, Nericell uses threshold based heuristics to find the road and traffic conditions.

Wolverine is a method which monitors traffic state by using the smart phone sensors. However,   for axes orientation we can find the horizontal orientation of the phone with the help of the magnetometer instead of waiting for a braking event.

We can reduce the energy intensive GPS usage and make the more reliable & efficient system by using this method. We compared energy consumption model for Wolverine with Nericell and decrease in energy consumption is shown and the battery life can be increased by using Wolverine method. To determine the traffic and road conditions, we will use machine learning techniques like K-means clustering and Support Vector Machine (SVM) instead of threshold based heuristics. Threshold based methods are less versatile and robust than Machine learning techniques.