Intelligent In-Car Navigation System
Proposed System
The proposed system would consist of few more fields with the available map attribute data. The fields would be used to store the speed of the vehicle in that interval of time. The speed of the vehicle would be captured from the GPS or Speedometer and it will be stored in the corresponding field.
| Id | An Unique id to refer to each segment |
| Speed (6 AM -7 AM) | Speed data captured in this time interval. |
| Speed (7 AM -8 AM) | As described above |
| Speed (8 AM - 9 AM) | As described above |
| ………………….. | …………………. |
| ………………….. | …………………. |
| Speed (11 PM – 12 PM) | As described above |
The proposed system consists of three systems:
- Map matching
- Collecting and Storing the GPS/Speedometer Speed
- Optimum routing based on the dynamic speed information
Map Matching and Storing Speed data
The following flowchart describes the work flow in collecting/storing the speed in the attribute table. The map matching algorithm would be implemented in such way that, it will find the vehicle’s position on the road of travel. The GPS time or system time would be used to retrieve the speed stored in the corresponding field of the attribute table. A tolerance limit would be used to check/filter whether the current speed should be updated in the attribute data or not. The tolerance would be used to eliminate the unexpected speed data say, at the time of traffic jam, the speed would be near zero. At that time, this speed data should not be stored. This process will be repeated, whenever the vehicle is in motion.

Figure 3
Optimum routing
When a user requests an optimum route by keying origin and destination(s), at the time of 10.00 AM, the road length and the dynamic speed stored (in the field 10.00 AM to 11.00 AM) in the attribute against the request time range would be retrieved and the travel time would be calculated. The optimum route would be calculated based on this dynamic speed data (say the output as dynamic route1).
Similarly, when served the same request at 8.00 PM, the road length and dynamic speed in the attribute data (field 8.00 PM to 9.00 PM) would be used for optimum route analysis (say the output as dynamic route2). Conventionally, a Navigation system would deliver the same result at both instances. In real time scenario, the output should be different which is addressed by the proposed system. The proposed system has incorporated the dynamic changes, as the speed data used in both the queries are different. Finally, the proposed system would produce an output more closely to the real time scenario. A simple flow diagram of the optimum routing of the proposed has been given.

Figure 2
Limitations
The system depends on the positioning accuracy of the GPS. In this segment, there are more chances to identify the wrong or nearest road segment instead of the right one. The multi-path and high rise building would also affect the accuracy. Since Galileo is to be launched in the near future, the above limitations would be reduced.
In case of a traffic jam, the speed of the vehicle would be nearly zero. It may not be a regular occurrence or one-off incident where the vehicle speed is not the true representation for a long term basis. Hence, the value should not be used for storing the attribute data. An option could be provided to the user, to inform the system, whether the speed data should be included or not.
Benefits of the system
In the developed countries, the sign board along the roads show the time required to reach the predefined destinations. But, this information could not be used for optimum route analysis. Because, the information would be the travel time needed to reach the airport, railway stations etc. The driver/user would need the optimum route between origin and the destination(s). At present in the developing countries, there is no available system as described above. The proposed would address these issues efficiently, because there is no need of traffic congestion details from outside the system.
We have seen that, the market available navigation system gives solution of optimum route with the static travel time stored in the map data. The proposed will help to improve the navigation system and to act with the real time data. In the area of fleet management and logistics, this system would support planning their services.
Conclusion
We have detailed discussion on calculation of speed in the GPS, accuracy of the speed from GPS, comparison of GPS speed and Speedometer, and its usage in the proposed system of the In-Car Navigation system. The proposed system will improve the geo-spatial applications in telematics. The proposed system is an important alternative to the existing static systems.
References
- Witte TH, Wilson AM. “Accuracy of non-differential GPS for the determination of speed over ground”, 2004
- Massimo Dragan, Michele Fernetti and Vassilis Spitadakis “Geographical Awareness for Modern Travellers: A GIS Application for Maritime Transportation in the Mediterranean Sea”, 2004
- Street, S. “Dijkstra’s Algorithm. SCC181 - Data Structures and Algorithms”, 2005