Abstract:

Context Aware driving assistance system (DAS) support is needed in potentially dangerous situations, especially for inexperienced drivers. The project motive is to develop a Context Aware Driver Assistance system that reduces the road accidents rate every day. In this paper we discuss about related technologies which generate warnings and alerts to driver continuously assisting him according to context which are integrated in Vehicle and Vehicle 2 Driver (V2D) ,vehicle to system for communications for assisting driver by HCI (Human Computer Interface).

The system will be adaptive to the environment changes and behaves according to the situations for maintaining a safe distance between vehicles, intelligent speed adaptation, remote diagnostics, driver drowsiness detections, night vision, automatic parking The project is involved with necessary modules like Context Sensing i.e. XML files of sensor,radar,lidar inputs. Context Acquisition Fusion(storing data ), Context Reasoner Engine (Rule-Base) i.e.Ontology protégé owl and Agents (API) i.e. java GUI, which connected to driver vehicle interface for alertness.

The inputs to the system are given by the XML files and are parsed by SAX parsers and according to the Context sensing system they are feed to Context Acquisition Fusion. The ontology modeling is used to define rules and restrictions for developing the system, which uses a reasoner called Racer Pro for reasoning the defined rules for related contextual situation.

                        An OWL-based context-model for abstract scene representation of driving scenarios is developed in the Protégé OWL ontology modeling tool. The OWL allows the classes to be classified and check taxonomy. We integrate scene-descriptions with a logic-based reasoning system Racer Pro, based on a set of transformation rules in Context service Rule Engine.

Problem description:

This system defines the driver assistant facilities to communicate, provide safety messages like warnings, alerts, directions, notifications  for in vehicle, vehicle to vehicle, vehicle to other system according to the context.

1) For in vehicle,

 The subsystem to be used for generating alerts and warnings are lane keeping, lane changing system, fuel economy, weather conditions, speed limit detection, rear vehicle detection, travel destination distance, night visions system, parking assistance and collision warning. The behavior of the system is according to weather conditions and sensor, radar, lidar, camera inputs of the system which is in the form of XML inputs stored in database. It warns the driver about the traffic, environment and alerts driver by taking rules and restrictions written in ontology protégé owl.

2) vehicle to vehicle in simulation GUI

In scenario let us consider 4 to 5 cars are moving on road and 2 cars got collided and accident is occurred. Thus a warning message with warning beep sound should be given to all other cars in the scenario apart from accident prone cars.

3) Vehicle to other system

Her two systems should be connected by Bluetooth ,and a scenario  is same as in vehicle to vehicle  on one system but a pop up notification that “accident occurred ,emergency visit required” with beep sound should be generated in the second system until we close the notification.

Tools:

1)    Protégé (3.4 or above version) OWL-s,Racer pro or Palette

2)    Sax Parser for parsing XML files

3)    Java (net beans or Eclipse)

4)    Database algorithm (any) for retrieval of data

5)    Database   oracle 

Back End:

Ontology Protégé OWL-S:

The rules and restrictions for the control systems subsystems(lane keeping, lane changing system, fuel economy, weather conditions, speed limit detection, rear vehicle detection, travel destination distance, night visions system, parking assistance and collision warning) are written in this protégé owl from which according to the environment given in the GUI is compared to the database and the rules are fired to produce the alerts, messages and warnings according to the situations and the control systems behavior which generates beep sound along with alerts or messages or warnings using reasoner plug in like racer pro or palette or any other .

Database:

All  the required regular behavior of the driver ,car ,environment is stored .The data is stored in the database for all the control systems and the data is compared from the database generated in the front end and generate the appropriate alerts, warns, messages and voice alerts. While comparing the data an optimized data mining algorithm (any efficient algorithm) is used to generate the feasible rule optimizing the data content from database.

Front End:

Java:

The GUI should be a car and a driver driving on road in 2 environments along with  traffic, atmospheres below:

1) Sunny

2) Dark situations.

The design should contain options like Interfaces (in vehicle, vehicle to vehicle, vehicle to system),a road scenario for 2-way and 4- way , a ‘start’ option, environments mentioned above(2 in number) ,stop option, text area showing the status of the car, and messages right below as pop up alerting the driver about the surrounding obstacles, maintain safe distance, speed limit when moving in over speed, available parking area sufficient for parking, messaging about fuel economy.  If road scenario is 2-way,it should contain some no of cars ,pedestrians, moving on the road. when start button is pressed it should ask for 2tasks  :

1)which car to be highlighted   and accordingly the subsystems like lane keeping, lane changing system, fuel economy, weather conditions(rainy or cloudy), speed limit detection, rear vehicle detection, parking assistance and collision warning  should generate warnings and alerts accordingly taking rules and restrictions written from ontology developed in protégé owl  to that car.

2)The GUI of movement of cars and whole process should be done dynamically and automatically immediately after selecting the car to be highlighted  for observation and warning and alert messages should be generated accordingly by the scenario in GUI, and  the status information generated on the text area. This should be saved in database with, at what time message is generated, which subsystem generated, speed of the car, and distance from that car to other obstacles etc, and the other details in the database for later verification after the process completed. The project should implement the above scenario and accordingly generate alerts, warning messages, voice beeps that warn the driver considering the control systems below: