An iNTELLIGENT Driving System.
Our project is to create an automated four wheeler driving system without using GPS for position tracking. . This is to reduce the occurrences of road accidents and make traveling safe. It integrates various technologies and provides a platform for a wide range of applications too. In the play, to detect obstacles we use sensors. For sensing other vehicles we use a reader which reads the signals transmitted from those vehicles. We use a roadmap to travel, and the automobiles’ dynamics to track its own position. This is used as an alternative to GPS. All these data are then combined and plotted on a frame buffer. Then we apply algorithms to generate path by considering the frame buffer. Using this we can achieve a safe and easy travel. We have simulated the working of i-Drive. It is a simple simulation showing the responses of the system when it comes across an obstacle or a vehicle approaching it.
The main cause of accidents is the long travel which results in loss of concentration and fatigue. The proposed system can thus reduce the burden of human involvement in vehicle driving. Though this system will work on all roads right now most optimal performance can be experienced on highways. It is not necessary to rely entirely on the automatic driving system. The user can also drive the vehicle manually if he/she wishes. This automobile runs on two modes: 1) i-Drive Mode (iDM) 2) Manual Driver Mode (MDM) The iDM is the auto driver mode and MDM is the manual driving mode. The user can switch to any mode anytime. Another great advantage of this system is that it doesn’t use any kind of telecommunication system and hence avoids the need for any network consequently reducing the communication cost. Some of the consequent applications of this system are: 1. Vehicle Tracing 2. Automatic Parking 3. Automatic Toll Paying
So, fasten your seat belts and get ready to take a big leap in the automobile technology!
The design of our system requires some of the existing technologies to be integrated. Each technology is incorporated in different components. These components are described below:
1. Presence Maker 2. Environment Perceiving Unit (EPU) i. Vehicle Spotter ii. Obstacle Detectors 3. Digitized Road Map (DRM) 4. Position Tracking Unit (PTU) 5. Scene Plotter 6. Smart Algorithms
EPU Vehicle Spotter DRM PTU Obstacle Detectors
1. Presence Maker It is an affordable transmitting device which is to be attached to every vehicle. The optimal range can be fixed as 50m. The function of this device is to transmit the following data continuously: Tag ID (Unique) Vehicle Dimensions (Length & Breadth)
For this we can use devices like Passive Read Only Memory RFID tags. These tags are capable of transmitting data written into them continually. 2. Environment Perceiving Unit (EPU)
This unit is responsible for sensing various entities around the automobile. The entities might be anything, from a piece of rock to a moving vehicle. It has two components viz. Vehicle Spotter and Obstacle Detector each detecting a class of objects.
This unit reads the signals transmitted from Presence Maker of vehicles, within the effective range and computes the relative position of the vehicle from ours. The data is then extracted from these signals and processed. Along with Unique Tag Id and dimensions the processed data contains following information: i. Position: The Receiver (e.g.: RFID reader) reads the signals and finds the relative positions.
ii. Speed & Acceleration: a) It can be calculated by comparing two samples of signals of a vehicle. b) A negative speed indicates that a vehicle is approaching ours’ whereas a positive speed indicates that a vehicle is moving away from ours’
c) The speeds are relative to ours’. d) A negative value of acceleration indicates deceleration and a positive value represents acceleration. Obstacle Detector:
These are usual sensors placed on front portion of the cars to detect obstacles. Each sensor is fixed with a particular orientation to monitor a wide range of objects. They prove to be very useful in emergency situations. We can use any proximity sensors. Then the processed information from the components described above is sent to Scene plotter.
3. Digitized Road Map (DRM)
This is a high detail data about the roads or various routes in an area. It is stored on a removable media. To make fast and efficient computation the data can be transferred to a flash memory or a RAM rather than accessing it always via a removable media. The updated version can be made available on the internet. The data from the above described components is then sent to the Scene Plotter. This is a simple illustration of DRM.
4. Position Tracking Unit (PTU)
This unit senses the direction and the displacement in that direction and updates the position of the vehicle with respect to the previous position. It uses an angle measuring device fixed to the steering shaft to measure steered torque. It has a digital speedometer to measure speed. The calculation of positions is done as follows:
Notations: a) Distance Traveled – d b) Speed - s c) Steered Angle (Torque) – d) Turned Angle – e) Sampling Interval – t f) New Position – NEWPOS e) Increases g) Old Position – OLDPOS Dynamic Variables: a) α b) Sampling Rate α s c) i.e. t α 1/s d) (i.e.), Sampling Rate
α 1 / Speed
Turns.((i.e.),decreases sampling interval)
Formulae: a) NEWPOS = s*t*sin b) Since s*t=d
After calculating the new position it sends this information to the scene plotter.
5. Scene Plotter The scene plotter uses a kind of Painters’ algorithm to plot the scene. Scene plotter has a frame buffer which is a two dimensional array of bits. A one (1) represents an obstacle or invalid path or another vehicle. Thus zeros represent a valid obstacle free path to move on. It uses information from EPU, PTU, and DRM and plots a frame for current scene. The algorithm used to plot the scene is described below:
Algorithm: 1.Reset all bits in frame buffer. 2.Using PTU we find the self position on the road map and plot that portion of DRM which is realizable by Vehicle Spotter on the frame buffer. The points in the range not on the road are considered as obstacles. 3.We plot the self dimension on the centre. 4.From the EPU we get the position of the obstacle, position and dimension of each vehicle and correspondingly plot it on frame buffer. 5.The main job is to map the real position to the corresponding pixel position which can be done by a mapping function as follows : Constants: i. ii. scaleX=maxX(frameBuff)/range(reader) scaleY=maxY(frameBuff)/range(reader)
Where, maxX - maximum number of bits on horizontal axis. maxY - maximum number of bits on vertical axis. range – the maximum distance the reader can sense.
Automated Vehicle Obstacles Free Path
Free Path Obstacles
This frame is then used by the Algorithms to predict a path to move on.
6. Smart Algorithms:
These are coded modules responsible to drive the vehicle and ensure safety. All units run independently. The Smart algorithms use all of them and make moves by passing control to the controllers.
These have many algorithms triggered when required. i. Scene Analyzer ii. Path Finder iii. iDM iv. MDM
The working of the above mentioned algorithms are described below.
Only the reset bits on the current screen indicate a valid path to move on. The algorithm scans the frame considering the following things: a. Speed and Acceleration of other vehicles. b. Safety which includes collisions or other accidents. c. Traffic rules.
The results of above consideration generate any one of the following four states:
a) Safe for iDM: There is a valid safe path to travel and the current situation can be handled by the system.
b) Safe for MDM but not for iDM: The current situation is too complex like a traffic jam but not so unsafe that a user can’t handle.
c) Not safe for MDM: There is a chance for minor problems.
d) Red Scene: There is no possibility to escape. An accident is going to happen and there is no chance for the user to handle it.
These results are then used by the executing routines which might be either iDM() or MDM() to decide the next action. The operations to be performed for analysis are simple instructions like AND, OR and XOR, as only bitwise comparisons are required.
Path Finder: It is a sequence of prioritized steps looking forward for the following in the given order: a) Safety. b) Traffic Rules. c) Steady and fast.
Algorithm: 1. 2. 3. First it looks for a free straight path and chooses it if it is free of accidents. If the vehicle in front of it is maintaining the correct speed limit it follows it. Else it goes to the faster lane and tries to overtake the vehicle if it finds no possible collisions while overtaking it.
When this mode is activated the user is allowed to control the automobile but the MDM module is activated. The working of MDM module is described below:
Algorithm for MDM: 1. Analyze scene in scene plotter. 2. Check whether the situation is a very severe one. 3. If so, activate safety tools such as balloons. 4. Else, check whether iDM is enabled. 5. If so, switch over to iDM. 6. Else, go to step 1.
Flowchart for Manual Driver Mode:
Analyze Scene in Scene Plotter
Is it a RED SCENE?
Is iDM activated?
Blow Up Balloons
Algorithm: 1. Analyze the scene in scene plotter. 2. Check whether it is safe for iDM to handle. 3. If yes, move on the computed path and Go To Step 1. 4. Else, check whether it is safe for MDM. 5. If no, Go To Step 1. 6. If yes, recommend MDM. Check for user response. a. If MDM is activated Switch over to MDM routine. b. Else check whether it is a Red-Zone. 7. If no, go to the corner and stop the vehicle. 8. Else take precautionary measures such as inflating balloons.
Flowchart for iDM:
Analyze Scene in Scene Plotter
Is it safe for iDM?
Is it safe for MDM? Yes
Move on the calculated path
Is MDM activated?
Is advice accepted?
Is it a RED Scene?
Go to the corner and stop.
Blow Up Balloons
1. Eliminates use of telecommunication systems, there by reducing the design cost considerably. 2. Flashing of head lights by vehicles approaching ours’ from opposite direction causes temporal blindness, leads to accidents in many cases. There is no such problem in our case. 3. In steep curves a human cannot sense the vehicle coming from the other side which may result in an accident. Our system can easily sense the presence of such a vehicle because it will definitely receive the signals from that vehicle.
Although useful, the functioning will not provide fruitful effect in crowded traffic. This can be achieved only when all the vehicles are automated. Then irrespective of the traffic the system is accurate.
Conclusion and future work:
The use of i-Drive will surely reduce the occurrences of road accidents by a considerable value. To cope with the exponential growth of vehicular traffic, it is necessary to adopt systems which have functions similar to ours’. But the drawback of current automating solutions is the usage of expensive systems and technologies like GPS. In the future, the wide acceptance of our technology will create an “Accident Free World”.
Digit May 2004 Edition - RFID Technology Interactive Computer Graphics by Donald P.Hearn – Frame Buffer http://www.HowStuffWorks.com – Auto gears and Passive RFID tags
The simulation is a limited illustration of the proposed system. It has a Status Notification area which explains in simple text, various aspects of the situation. A screen shot of our simulation is shown below: