Autonomous Drone Navigation in Closed Environments Using SLAM Techniques

The improvement of independent navigation structures for drones has revolutionized severa industries, from agriculture to look and rescue operations. however, navigating drones in closed environments, consisting of warehouses and factories, poses unique demanding situations due to the limited outside cues and limitations. In this paper, we suggest a approach for the problem of self-enough drone navigation in closed environments using Simultaneous Localization and Mapping (SLAM) strategies. SLAM is a well established method for robots to construct maps of unknown environments while simultaneously estimating their place within the map. inside the context of drones, SLAM algorithms use facts from onboard sensors, cameras, and LiDAR to create a map of the surroundings and track the drone’s role and orientation in actual time. This document is then used to plan a safe and green path for the drone to navigate via the surroundings. Our proposed tool uses function-based total SLAM and occupancy grid mapping strategies to gain robust and correct navigation in closed environments. characteristic-primarily based SLAM techniques extract and music extraordinary functions inside the surroundings to assemble a map. on the equal time, occupancy grid mapping strategies use probabilistic fashions to represent the occupancy of the environment. to evaluate the overall performance of our machine, we conducted experiments in simulated closed environment and assessment our effects with ground truth statistics.