Intellectual transport system for human safety using machine learning approach

Abstract

Generally, it is projected that a combination of automated and manually vehicles would be core of the intellectual transportation scheme. As a result, resolving the unsafe conditions raised by such a combination of automated and manually cars are critical until automated driving becomes widely used. Automated driving have concerns including a poor intent detection accuracy and inadequate genuine performances while forecasting drive orientation even as intellectual infrastructure has become more complicated; such issues have a significant impact on the safety and pleasure of heterogeneous traffic network. As a result, automated automobiles' capability to forecast drive directions in live time depending on the traffic condition must always be enhanced, and scientists should seek to develop a much more sophisticated intellectual system. In this research, we present machine learning-based highway safety solutions for a 6G-enabled intellectual traffic system with a mix of automated and manned automobiles. A drive trajectories database and an instinctual database are used as networking feeds to great memory neurons in the 6G-enabled intellectual traffic system in this system, with the max pooling calculating the parameters of the every intent. The ultimate intent likelihood would then be calculated by combining the average rule with the choice level. The suggested approach yields intent detection accuracy of 91.58% for left changing lanes and 90.88% for right changing lanes, correspondingly, leadership has a significant both precision and genuine intent detection and addressing the lane - changing issue in a congested traffic scenario.

Keywords

• Automation
• Transportation
• 6G-enabled
• Intellectual
• Traffic-system
• Big data