Implementation of Steering Control System for The Prototype of Autonomous Car Using Raspberry Pi

In the advanced period the vehicles are engaged to be robotized to give human loose driving. In the field of car different viewpoint has been considered which makes vehicle robotized. In this project thinking about the distinctive highlights what more the cost on a little scale a three-wheel vehicular mechanical model has been outlined that will take after the path and evade deterrents. Self-ruling autos are a creating innovation which may end up being the following enormous development in individual transportation. This report starts by depicting the scene and key players in the self-driving auto showcase. Current capacities and also restriction and chances of key empowering advances are looked into alongside a talk on the effect of such advances on society and the earth. Most effect including diminished movement and stopping blockage autonomous versatility for poor individuals expanded security and vitality preservation and contamination decreases may be noteworthy when self-sufficient vehicles end up normal and reasonable to average folks. Raspberry pi is the focal processor of our autonomous auto. Different pictures are caught by the camera module on this pictures different image handling methods are utilized to accomplish artificial intelligence.

Traffic light and Sign detection assist autonomous locomotives in industries by providing required commands for facilitation of flexible manufacturing system. The autonomous locomotives in industry are used for material handling. Automatic sign recognition guides autonomous locomotives to locomote in proper direction. The path tracking of autonomous locomotives is described by steering control system. The main objective aims, design a method for steering control system for autonomous locomotives by reorganization of sign and traffic lights. The system provides efficient locomotive system in flexible manufacturing environment. Image processing techniques are employed for regulating traffic signs and to command certain actions. The input to the system is video data which is continuously captured by web camera interfaced to raspberry pi through open cv platform in which raspbian os is used. Images are pre-processed with several image processing techniques such as hsv color space model techniques is employed for traffic light detection, for sign detection against hsv color space model and contour algorithm has been used. The signs are detection based on region of interest (ROI). The ROI is detected based on the features like geometric shape and color of the object in the image containing the traffic signs. Steering control system uses dc motors and motor drives for functioning.

Gurjashan Singh Pannu et al, proposed a “Design and Implementation of Autonomous Car using Raspberry Pi” the summary is as follows,

Elodyne 64-beam laser produces a detailed 3d guide of the surroundings. The auto at that point joins the laser estimations with high-resolution maps of the world creating diverse kinds of data models that enable it to drive itself while maintaining a strategic distance from deterrents and obeying traffic laws. Components utilized for designing Google car are sensors four radars mounted on front and back bumpers 1 camera situated close to the rearview reflector GPS wheel encoder that decides vehicles position and monitoring of movements lidar velodyne 64-beam laser produces keen 3d guide of the surroundings.

A new traffic sign recognition system has been displayed in this paper. The application software created in this work perceives and classifies traffic signs from an input image. The image processing techniques utilized as a part of this product incorporate a preprocessing stage regions of interest detection potential traffic sign detection as per the traffic sign shape patterns lastly the recognition and classification of these potential traffic signs as per a database of traffic sign patterns. The execution of this application relies upon the quality of the input image in connection to its size difference and the way the signs show up in the image. With this thought the rates of perceived signs for this application are high. As further work a neural system could be actualized so as to acquire all the more precisely the observational parameters utilized as a part of the application. Besides, the application could be enhanced by actualizing inserted equipment for use in dynamic applications.