Intelligent Transportation Network Model for Drivers Based on Observed Patterns

Drivers' erratic behavior is increasingly becoming a threat to road safety. Indeed, such activities are very dangerous to the rest of the parts of the vehicle, and might lead to severe accidents. In this regard, this research introduces a Driver Decision Support System within Intelligent Transportation Systems in order to detect and manage unusual driver behavior. DDSS works with a reliable method for the prediction of driving behavior. In order to be effective in ITS scenarios, the system has to identify unusual driving patterns consistently and issue a warning to vehicles in the vicinity to change lanes or reduce speed. K-Means is one of the algorithms used for predicting driver actions in this system. To prove the effectiveness of the algorithm, a comparison has been done with Improved Logistic Regression, Naive Bayes, Tuned K-Nearest Neighbors, Random Forest and Support Vector Machines. If the DDSS is integrated into the ITS infrastructure, accident prevention efforts can be enhanced to a large extent. It aids in monitoring and analyzing driver behavior and provides timely interventions that enhance safe driving practices, thus reducing accidents on the roads. This work therefore contributes towards a more secure transportation system by reducing the rates of accidents by a considerable margin due to reckless driving. The DDSS suggested in this research has very promising potential for enhancing road safety and preventing accidents with its innovative way of predicting and managing driver behavior. These results of the in-field trial represent a very effective and reliable algorithm predicting driver behavior.