Design and Development of IOT, Web-Server and ML-AVPR based Intelligent Humanoid Robot for Traffic Assistance

摘要

In India, there is lack of traffic assisting persons (traffic police). They are basically engaged in various activities by authorities. Major problem facing is while traffic is switched OFF because of technical problems like battery backup, under maintenance condition or many more. To overcome these problems, the intelligent traffic assisting robot can be used at the square of traffic signal. When power ON, it will act as intelligent monitoring and alerting system. When traffic signal goes OFF, it will take control of signal rotating with its basic task of intelligent monitoring and alerting. Humanoid robot is designed with low power devices that takes very less power than that of required for signal operation at the time of mains power OFF. By same propose system, both manual traffic like control (by robot) and signalling control may be performed at a time if required. Intelligence required in the proposed system is for identifying specific instance as traffic breaking. Various instances may be monitored using back end program designed. But here, traffic breaking is discussed in details. With traffic breaking instance, intelligent algorithms perform tasks as identifications of traffic breaking vehicle, driver and recognition of number plate for assigning fines to particular. It is performed using machine learning model called ML-AVPR Algorithm. A novel approach for image segmentation of vehicle area, driver area and licence plate area is proposed for better analysis through Machine Learning algorithm. The segmented images are given as input to the Machine learning algorithm that identifies various required parameters in real time manner. The results are satisfactory and appeared in real time sense.