Real time Accident Vehicle Detection using GPS and GSM

Project Title

Real time Accident Vehicle Detection using GPS and GSM

Project Area of Specialization Electrical/Electronic EngineeringProject Summary

In this project, we have avoided the false alarm situation caused for some conditions, increased the accuracy of accident detection using more than one sensor, cut the project cost by using the already existing infrastructure available in the victim’s mobile phone. To avoid the false alarm we have one manual switch in the vehicle itself which must be pressed within 10 second of false accident detection and hence avoiding any false intimation. We are using front bumper sensor, position encoder along with the accelerometer sensor in order to increase the accuracy of accident detection. Bumper sensor will tell the microcontroller how much force/pressure has been applied on it and its obvious the pressure will be more in case of accident. Position encoder is used for calculating the speed of vehicle and it is expected to change drastically when accident being met and adding another layer of reliability. The accelerometer sensor as usual tells the microcontroller if there is sudden change in the acceleration. Nowadays every android phone have inbuilt GPS, GSM modules which we are using in order to get the accident spot location and to send the SMS.

Project Objectives

1. To detect the accident with accuracy and certainty in easy way.

2. To find out where exactly the accident met.

3. To save valuable lives by using modern technologies.

4. To alert the emergency services whenever accident met.

Project Implementation Method

Our signal processing goal is to classify the original data into two classes, fall and not fall. In this system, the input data from 3 axis accelerometer was kept and processed in real-time with sampling rate of 60 Hz or higher. The signal from MEMs accelerometer was converted by 10 bits ADC into integer range between 0 and 1023. The sensor was embedded in a vehicle seat to fix the accelerometers axis so that the response of acceleration data is well defined. The classification of the fall detection utilized the 3-axis acceleration signal from MEMs accelerometer and the ground speed from GPS module. In general, vehicle fall can be classified as linear fall and non-linear fall. The linear fall is concerned about fall without external force, which is free falling condition that only z-axis acceleration changes. The non-linear fall occurs by the external force. The nonlinear fall detection is decided by all 3-axis acceleration data from accelerometer and the ground speed from GPS module. To determine the accelerometer output, two frames of acceleration data, which include 3-axis acceleration at present time (t) and prior time (t-1), are used for analysis. For a linear fall, the z-axis acceleration follows free falling condition which is given by where the AZ is the z-axis acceleration. In a non-linear fall, two frames of acceleration data are used. From non-linear fall experiments under most likely situations, we found that the change of acceleration between two consecutive frames should be more than 15.5 mls2. Thus, the non-linear fall condition is given by Where the An t is acceleration from x, y or z coordinate at the present time frame and An t-l is acceleration parameter from x, y or z coordinate in the previous time frame. From equation, if the difference of acceleration in two time frames is more than 15.5 mls2, the first condition of non-linear fall accident of vehicle is met. The ground speed from GPS module then used to decide whether actual non-linear fall accident occurs. If ground speed becomes zero after detection of large acceleration change as indicated in equation (2), non-linear fall detection in vehicle is detected and fall alert message will be sent. However, false detection may occur in case of a severe brake because data are not kept and processed over a long time frame. Normally, there is noise in z-axis while vehicle rides over knotty surface. The noise is filtered by averaging acceleration data all of three axes over five time frames. The fall detection and alarm system for vehicle accident operates according to the flowchart as shown in Fig. 2. After system start, microcontroller periodically gets 3-axis acceleration data from the accelerometer. If acceleration in z-axis satisfies the free falling condition in equation (1), a linear fall is identified. If the system detects a linear fall, the position of accidental place will be saved and sent via SMS

Benefits of the Project

1. This system can be implemented in all of the vehicles.

2. This system can be implemented in industries.

3. Wherever security is important, this system can be implemented.

4. It can be used to control the accidents.

5. By implementing this project, many of the precious lives can be saved.

6. Low power hardware components being used in our system.
7. Uses some already existing hardware components of mobile phone hence lower the total cost/budget involved.
8. Use of more than one sensor increases the accuracy of our system.
9. False alarm switch can avoid any false intimation hence add more towards the reliability

Technical Details of Final Deliverable

Hardware used-

1. Microcontroller-

We are using AT89S52 controller.

2. LCD-

16x2 LCD is used. LCD displays latitude and longitude values of location.

3. GPS -

GPS satellite transmits data that indicates its location and the current time. GPS continuously sends latitude and longitude values to microcontroller.

4. GSM -

GSM SIM 300 is used. GSM receives co-ordinates from microcontroller and sends message to mobile number store in our system.

Software tools

Arduino. .

Final Deliverable of the Project Hardware SystemCore Industry ITOther IndustriesCore Technology Internet of Things (IoT)Other TechnologiesSustainable Development Goals Good Health and Well-Being for People, Decent Work and Economic GrowthRequired Resources