Design of an AI Base Framework for Speed Limit Prediction and Control Using GPS Data

Computer Science

Road transportation and automobile requirements are on a daily increase and are directly proportional to the frequency of accidents. Therefore, developing an intellectual automobile system and other precautionary measures to safeguard drivers have become imperative. One of the important components to deal with is the speed limit signs (SLS) for the realization of an improved advanced driver assistance system (ADAS) with fully restricted and intelligent speed limiter systems. The existing optical recognition systems involve scanning speed limit signs to recognize the acceptable speed limit in a specific location. This can restrict the use of such ISA technology to only areas with frequent speed limit signs. To address this problem, this paper explores the potential of an artificial intelligence algorithm to accurately predict the speed limit of different streets from the speed limit datasets obtained from GPS devices. The GPS data is recorded over the test track to form a comprehensive dataset for the vehicle track. The datasets are then used to train a neural network model that can predict the location's speed limit, which is then communicated to the vehicle ECU offline. In addition, the trained model can be used to provide the adjusted speed limit as input through the MCU and then apply the maximum speed restriction to control the vehicle from overspeeding. Therefore, the vehicle can set the vehicle using the predicted speed limit. Experimental result in MATLAB 2021a shows that the NN performs well in predicting the vehicle's speed limit. However, the model mean square error for the validation set was 92.33% which is the lowest when compared to the testing and training state. after achieving the research aim, the model can enhance safe driving in Nigeria.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.