Robotic navigation in the presence of static and dynamic obstacles

Abstract

Computer Vision is the analytic study of motion dependent machines that extract valuable information from an image and perform some algorithmic processing on the captured images to derive necessary information to solve a particular or a set of tasks. Computer vision is the reconstruction of explicit, meaningful descriptions of physical objects obtained from their images. As far as scientific discipline is concerned, it moreover concerns itself with the theoretical aspects related to artificial intelligent systems that extract valuable information from the captured images. The role of computer vision in automated system is providing the detailed information about the working environment. A robust vision system should always be able to detect as well as identify objects reliably and provide an accurate and precise information as well as representation of the environment conditions to high level processes. The robust vision system should be highly effective and efficient, allowing all the objects which are limited and used as resource to respond quickly to a changing environment. Computer Vision is the key aspect to the design of the automatic responsive system by virtue of which the objects in the system localize themselves in the environment and act according to the changing nature of the environment variables.
The objective of the project was to build an autonomous vehicle for urban road which could guide itself to the specified location based on its decision making quality on basis of the path analysis using image processing and GPS. It would guide from the present source to the destination based on the path created using the wave points with obstacle avoidance of all objects, both static and dynamic in nature. It should also be capable of switching into manual mode control upon instruction being issued i.e. it can be controlled manually using the manual wireless control using X-Bee wireless module.