Abstract: This project is developed for detection of obstacles coming in path of the vehicle. It is designed keeping in mind as an aftermarket product. The notification inside the vehicle is designed to warn the driver without being distracted. The controller actuates an alarm when the sensed distance of the object is less than the safe distance. Collision Avoidance system is designed for the prevention of front-as well as rear’end collisions with vehicles, which are either stationary orin motion. Thus we have proposed a system which is suitable in various accidental situations.
Keywords: accidents, collision, distance measurement, driver, ultrasonic sensors.
Due to increasing population and lifestyle requirements, vehicles have become an object of daily need. However, today’s road situation is very bad and on an average, every minute at least an accident takes place causing casualties.One of the accident scenario is as shown in figure 1. Due to accidents, every year as many as 10 million people suffer injuries and out of them, two tothree million are serious cases.In India alone 1,20,000 people die per year  due to vehicle accidents. India tops in death due to road accident across the world.The hospital bill, damaged property and other costs are expected to add up to 3% of the world’s gross domestic product . With the aim of reducing injury and mishap severity, pre-crash detecting is becoming an area of vigorous research among government sector, automotive companies and researchers. Automobile accident statisticsindicate that the biggest threat a driver usually faces is from the other vehicles. Therefore, developing a driving assistance system,whichtargets at alertingthe driver about its driving environment and collision possibilities with other vehicles, has gained a lot of attention.
Fig.1: Vehicle collision
Different Collision Scenarios
Fig.2: Different Collision Scenarios
Above figure 2 shows, the various situations a driver faces in real time on roads. In the scenario (a), the collision is imminent to happen at a roundabout where the warning from sideways of vehicle is important to alert driver. In case (b), when the vehicle changes its lane the detection from rear side should be done. For situation (c), during overtaking, detection of vehicle coming from other side should be done. Also when sudden brake is applied there may be possibility that the vehicle coming from rear will collide with our vehicle, so adequate indication needs to be provided to alert driver of the accident.
There are a large number of factors that cause vehicle accidents such as inattentiveness, late reactance due to fatigue, stressful environment, consumption of alcoholic beverages leading to judgmental mistakes, environmental conditions like foggy, rainy, snowy weather leading to vehicle failure, incorrect road directions, sudden intervention of an obstacle in the path of the vehicle for e.g. animals and pedestrians and skidding of vehicle due to urgent breaking.
II. LITERATURE SURVEY
Various approaches have being adopted to avoid or reduce accidents e.g. fatigue measurement of a driver by constantly scanning his/her eyes and heartbeat response, vehicle communication system  in which each vehicle is equipped with a special device that transmits signals and surrounding vehicle system respond to it and distance is calculated using GPS System and accordingly warning is given. During night hours accidents occur due to poor visibility and limited coverage of headlights on curved roads, so a technology has been implemented by BMW in which head lights of vehicle sense the curve of roads and adjust accordingly, so as to give wider field of view to driver.
Currently car manufacturers are developing cars with newer and improved systems that focus on driver safety. We are proposing a system that would assist the driver inavoiding accidents or collisions.The aim of the project is to avoid collision between vehicles by providing auditory and visual alarm by detecting the obstacles. We are using ultrasonic sensors.Ultrasonic transducers are very accurate in distance measurement. They are also the cheapest solutions available. Ultrasound waves are useful both in air and underwater  conditions, and areadequately fast for most of the applications.The information supplied by multiple ultrasonic sensors can deal with the problem of spatial uncertainty of unknown, unstructured environments in several applications of advanced robotics, such as flexible Industrial automation, service robotics, and autonomous mobility  and in Defence for the detection of intruders.In simple systems, a low cost version of ARM processor can also be used to lower the system cost.
A lot of work has been already done related to distance measurement and vehicle collision avoidance. A list of some of the patents is given in Table 1. Y. Chey et al.  had designed a system that uses a an active or passive acoustic ranging device fixed on back side of the vehicle to automatically sense the existence and direction of objects in the surrounding area of the vehicle. D. Shaw and J. Shaw  implemented a collision avoidance system based on the use of two set of laser radars. They also used a wheel skidding detection system based on deviation between the directional change rate detected by a steering wheel rotation sensor and the actual directional changing rate predicted by a laser gyroscope. However, the system is weather susceptible and not useful in foul weather. J. Lemelson and R. Pedersen  have used a GPS fixed in the car to get an accurate position of a car. This position information is transmitted to a central server and the trajectory of a vehicle with respect to other vehicle is predicted and a warning is given if necessary. However, disadvantage is that each vehicle should be equipped with a GPS system and also the warning is not given with respect to stationary obstacle e.g. Wall, Barricades etc. P. Chang  has used a stereo camera for the prediction of vehicle path and detection of possible collision with obstacle. However, use of camera makes the system very complex and required the use of digital signal processors making the system costly.
Table 1: List of patents related to vehicle collision detection and avoidance system
S.N Patent Title Patented By Patent Number
P1 Vehicle detection and collision avoidance apparatus Young Chey, Irvine and Calif US 4626850
P2 Automobile collision avoidance system David Shaw and Judy Shaw US 5314037
P3 GPS vehicle collision avoidance warning and control system Jeromelemelson and Robert Pedersen US 6275773B1
P4 Stereo-vision based imminent collision detection Peng Chang and Theodore Camus US 7660436B2
A number of safety warning systems are currently available in the market. They are discussed below.
A. Sleepiness Warning System
Industrial Technology Research Institute has used the ultra-wide bandwidth technique to integrate low power pulsed electromagneticwaves to precisely measure the driver’s physiological signals such as heartbeat, respiration etc. Its main purpose is to emitwarning signal to awake the driver before falling intosleepiness and to avoid any possible accidents.
B. Low-Speed Following Mode (LSM) System 
The LSM exploits mm wave radar to detectthe acceleration, deceleration and stopping of the frontcar to estimate the distance of a vehicle from the front vehicle and also controls the brake and the fuel systems to maintain the vehicle position within the safety range. If the front car experiences an abnormal condition, the system will raise an alarm sound to warn the driver.
C. Lane Keeping Assist System (LKA) 
The LKA uses a camera to monitor the passing or dividinglane in the front and the system will givesa warningsignal if the driver crosses the passing lane or enters intothe opposite direction passing lane without giving a proper direction signal.
D. Intelligent Night Vision System (INVS)
During night or in heavy rain the driver’s visiblerange isrestricted,also the light illuminatingrange is narrow to allow the driver from clearlynoticing any pedestrian along the roadside.The INVS uses an infrared camera and displays front road conditionon the LCD screen in thecar.
E. Adaptive Front-Lighting System (AFS)
The AFS detects the car’s moving direction by using a direction sensor implemented in a wheel and then uses the unit’s control motor to rotate the headlight. Also with the installation of High Intensity Discharge (HID) Xenon light to extend the light illuminating range so asto enable the pedestrians and the obstructions in the vehicle direction to be visible to the driver.
F. Video Parking Assist System
This system uses camera to extract the images of the car in the back and side directions and in coordination with the collision avoidance radar system and thedynamic auxiliary linear system to inform the driver of the wheel rotation direction and angle to assist the driver tocomplete the reversing or parking action. Currently it is a new system, in coordinating with the wheel rotating mechanism, without requiring the direct involvement ofthe driver to automatically assist the driver to reverse the car into the garage or to park the car along the road side.
III. Techniques for Distance Measurement
This section deals with the different technologies used for distance measurement. The nature of the application rules out a physical contact with the obstacle. Some sort of wave must reflect off the object and be used for the measurement of the distance from the object.To detect the obstacle many technologies have been investigated to measure the distancesuch as radio waves, infrared waves, magnetic sensors etc.
Radio waves are electromagnetic radiationshaving wavelengths in the electromagnetic spectrum higher than that of infraredwaves. Radio waves have frequencies ranging from 3 KHz to 300 GHz. The problem with these waves is that the small obstacles would require a very high radiofrequency, resulting in a complex, perhaps power hungry circuitry. This ruled out the use of radio waves for object detection in our project.
For detection of obstacles use of infrared (IR) waves, is astrong possibility. Infrared detectorsare classified into two types, active and passive. Active infrared sensors operate by transferring energy either from a laser diode or a light emitting diode (LED). The reflected energy is concentrated onto photo detectors, whereas in passive infrared system energy emitted by the object in the range of visionis detected andwe have to usesignal-processing procedures to extract the desired information. The passive infrared systems can detect presence, of obstacle with low power requirements, and they also less complexity. This option is ruled out because of poor reliability in stormy weather and small range.
Magnetic sensors can also measure the distance between vehicles, as every vehicle has a magnetic field. The researchers have recently found that a vehicle’smagnetic field has an inverse relationship with small distances. The relationship has used for estimation of vehicle’s position using its magnetic field, when the vehicle is present at a distance less than a fewmeters. It can be useful for detecting impending collisions fair before they occur.Anisotropic magnetoresistive (AMR)  and sonar sensors are adopted for the development of a sensor system.However, thesesystems are very costly and more research is to be done in this field.
In the microwave radar technology, microwaves are used for the detection of obstacles. A drawback of this system is that on highways it will detect echo signals from cars in adjacent making it difficult in differentiating various echoes. Becauseof lowangularresolution of microwave sensors, thedirection of objects cannot be determined exactly. Also, if the objects are too close to each other they cannot be differentiated. Microwave radars also have difficulty in differentiatingradar signals coming from neighboring cars with analogous equipment. If there are more than two vehicles with the same radar apparatus on the same passage, the signalsturn out to be very confusing. Thus we are not using it.
Obstacle detection using audio waves is inexpensive, has the range we desired and functions well in poor conditions. Althoughit is susceptible to temperature, itcan be calibrated  in the programming. So it isreasonably the best available choice. Here, we have two possible approaches for distance measurement, the Doppler method and use of echo time. The Doppler method measures a change in distance requiring either the object or the sensor to be in motion. On the other hand, echo time can detect stationary objects as well.A comparison of various detection techniques is given in Table 2.
Table 2: Comparison of various technologies for the measurement of distance of obstacle.
IV. DESIGN AND IMPLEMENTATION
In this paper, we are focusing on front, rear as well as sideways collision probabilities.Fig. 3 shows a typical arrangement of positions of the sensors. Three sensors are used on each side of a vehicle to detect the obstacles in that direction.
Fig. 3: Typical placement of sensors in a vehicle in the proposed vehicle collision detection and avoidance system.
The block diagram of the proposed system is shown in figure 4.
Technology Advantages Disadvantages
Ultrasonic Compact in size, easy to install. Sensitive towards atmospheric temperature and turbulence due to air.
Microwave Doppler Operates well in foul weather. Vehicles speed can be measured directly. Cannot detect immobile vehicles or vehicles moving less than speed of 5 mph approximate.
Passive (receive only) infrared Wider viewing distance in foggy climate than with perceptible wavelength sensors Degradation due adverse weather conditions.
Active (transmit and receive) infrared Wider viewing distance in foggy climate, Directly calculate vehicle speed. Potential degradation due adverse weather conditions.
Fig. 4:Block diagram of the proposed vehicle collision detection and avoidance system
The detail of the implemented design blocks are as follows:
A. Ultrasonic Sensors
Ultrasonic transducers are based on the physical attributes and several other effects of ultrasound of a specific frequency.Sound waves are considered ultrasonic when they have a higher frequency than the human ear can detect (>20 kHz). It transmits and receives the ultrasonic waves of certain strength. There are two types of ultrasonic transducers available i.e. in piezoelectric and electrostatic versions. The piezoelectric type is preferred due to its simplicity in usageand lower cost .An array of ultrasonic sensors has been installed on front, rear and sides of the vehicle to measure the distance from obstacles.The velocity of propagation of ultrasonic waves in the air is approximately 340 m/s at 15??C of atmospheric temperature. The distance measurement of an object or obstacle from the sensor ultrasound module is governed by the medium and its temperature. The velocity of ultrasonic waves in the air is calculated as
v= 340+0.6(t-15) m/s (1)
Where, t is the temperature in degree Celsius.
The ultrasonic sensors detect both the moving and the stationary objects around the vehicle and will reports their absolute distance to the processor.
B. Temperature Sensor
Sensor TMP136 is the probable temperature sensor for calibration of the Ultrasonic sensor, as the Ultrasonic sensor is temperature dependent.
C. Microcontroller ARM Cortex M0
The system uses an ARM Cortex M0 microcontroller which is a most recent, advanced and economical microcontroller. It is the smallest ARM processor available.The exceptionally small silicon area, low power and minimal code footprint of the processor enables developers to achieve 32-bit performance at an 8-bit price point. Following are the features of ARM Cortex-M0:
- The smallest ARM processor- The code density and energy efficiency benefits of Cortex-M0 mean that it is a natural and cost effective successor to 8/16-bit devices in a wide variety of applications.
- Low power-Itconsumes as little as 16??W/MHz
- Simplicity- With just 56 instructions, it is possible to master quickly the entire Cortex-M0 instruction set and its C friendly architecture, making development simple and fast.
- Optimized connectivity- Designed to support low power connectivity such as Bluetooth Low Energy (BLE), IEEE 802.15 and Z-wave, particularly in analog devices that are increasing their digital functionality to pre-process and communicate data efficiently.
D. Indication Module
The notification inside the vehicle has designed to warn the driver without being distracted. The controller actuates an alarm when the sensed distance of the object is less than the safe distance. Audio and visual peripherals are used for indication purpose.
The Ultrasonic sensors are interfaced to the ARM Cortex M0 microcontroller, where the triggering and measurement is done using singleinput and output pin. The sensor will transmit an ultrasonic wave and produce output pulse that correlate to thetime required for the echo burst to return back to the sensor. The echo pulse width istransmitted to the central processor inside the vehicle. And also the system includes a speed sensor for sensing the current speed of the vehicle. The distance to target can easily be calculated from the distance measured from the data provided by the sensor and the current speed of the vehicle. The microcontroller computes safe distance to the object.
With the help of stellar table as shown in figure 5, controller will determinemode in which the vehicle is being operated and accordingly indicate the driver of his probability of collision.
Fig. 5:Stellar Table showing speed vs. safe distance
a) Advantage and Limitation of proposed system
- The proposed system has several advantages that include:
- Reduction in accidents due to driver’s negligence.
- Simple and easy operation.
- No remodeling of existing vehicle required.
- Less power required.
However, the proposed system also has following limitations:
- Interference with other vehicles using similar transmitting equipment.
- Poor reflection of ultrasonic waves (and hence the collision detection accuracyreduces) due to rough or angularly placed obstacles.
b) Future Scope
- Cameras and image processing system can be used to greatly improve the effectiveness of the proposed system.
- A GPS module can be used to get the exact location and position of vehicle and in case if accident happens at all, then anautomatic notification can be sent to the nearest emergency unit using a GSM module.
- A GPS can also be used to give warning at accident prone areas so as to alert the driver beforehand.
- The proposed system can be modified to assist in lane changing as well as in parking.
Vehicles collision avoidance system is very useful in today’s society.It makes drivers more aware of their surroundings.Therefore, it will be easier for them to avoid collisions. As highways, roads and parking lots are becoming more and more crowded with more people driving larger vehicles that are hard to maneuver, the proposed vehicle collision detection and avoidance system will be a boon for safety of passengers and vehicles.
 S. Ithadi,’Vehicle anti-collision device-Say no to accidents’,http://www.changemakers.com/entries/vehicle anti-collision device.
 W. Jones, ‘Building safer cars’,IEEE Spectrum, vol. 39, no 1, pp. 82’85, 2002.
 C.W. Hsu, C.N. Liang, L.Y. Ke and F.Y. Huang, ‘Verification of on-line vehicle collision avoidance warning system using DSRC’, World Academy of Science, Engineering and Technology 2009.
 J. David and N. Cheeke, ‘Fundamentals of ultrasonic waves’,CRC Press, Florida, USA, 2002, ISBN 0-8493-0130-0.
 A. Sabatini, V. Genovose, E. Guglielmelli, A. Mantuano, G. Ratti, andP. Dario, ‘A low-cost, composite sensor array combining ultrasonic and infrared proximity sensors,’ International Conference on Intelligent Robots and Systems, Pittsburgh, Pennsylvania, vol. 3,pp. 120-126, August 1995.
 Y. Chey,Irvineand Calif, ‘Vehicle detection and collision avoidance apparatus’, patent number US 4626850, 1986.
 D. Shaw and J. Shaw,’Automobile collision avoidance system’, patent number US 5314037, 1994.
 J. lemelson and R. Pedersen, ‘GPS vehicle collision avoidance warning and control system’, patent numberUS 6275773B1,2001.
 P. Chang and T. Camus, ‘Stereo-vision based imminentcollision detection’, patent number US 7660436 B2,2010.
 P.C. Philipp, E. Udo and U. Peter, “Experimental evaluation of eye-blink parameters as a drowsiness measure.” European Journal of Applied Physiology, Volume 89, Issue 3-4, 319-325, 2003.
 B. Beechie, J. Mueller and A. Varma, ‘Driver selectable low speed mode technology’, patent number US8485941 B2, 2011.
 S. Ishida and J. Gayko, Development, evaluation and introduction ofa lane keeping assistance system. In Proceedings of the IEEE IntelligentVehicles Symposium, pages 943’344, Parma, Italy, June 2004.
 Jones, D. Willie, ‘Safer Driving in The Dead of Night”, IEEE Spectrum.ieee.org. Retrieved 2008.
 G. Jenssena, A. Cato and K. Sakshaug, ‘Behavioural Adaptation to Adaptive Front Lighting Systems (AFS)’.SINTEF Transport Safety and Informatics.
 C. Pradalier, S. Vaussier and P. Corke, ‘path planning for a parking assistance system implementation andexperimentation’.
 S. Taghvaeeyan and R. Rajamani, ‘The development of amr sensors for vehicle position estimation’, American Control Conference, San Francisco, CA, USA June2011.
 Johan Huijsing, ‘Sensors and Actuators Integrated Smart Sensors’, Vol. 30, pp. 167-174, Jan 1992.
 H.He and J.Liu, ‘The design of ultrasonic distance measurement system based on S3C2410,’ Proceedings of the IEEE International Conference on Intelligent Computation Technology and Automation, pp. 44-47, Oct. 2008.
...(download the rest of the essay above)