Implementation of collision avoidance and path planner using SLAM algorithm in MATLA

Researchers are interested in developing robots that are able to make a 2D/3D map of an unknown room and sending data to the users. For a long time, localization of a robot, and at the same time mapping its environment, has been a central objective of the robotics industry. Simultaneous localization and mapping (SLAM) is a procedure by which a mobile robot can build up a map of an environment and at the same time use this map to calculate its own position. We designed an algorithm to find whether a SLAM algorithm can be used to avoid obstacle and plan a path to its target and simulated it in MATLAB. The simulation was able to track the landmarks in the minimum range and to find both the optimal path from starting point to ending point while it approaches its chosen target. This algorithm incorporates a nearest-localizing technique to find new landmarks with respect to existing landmarks. As a result the robot avoids any landmark if it is in its way thereby avoiding collision while planning the course to its target. In conclusion, this simulation indicates that a hardware implementation using high range laser sensors or using a Kinect system with a GPS module would allow a robot to create a map and find its target. This project supports the prospect of using robots in applications such as firefighting, military maneuvers, underwater exploration, and more.