Development of prediction based emergency obstacle avoidance module by using LIDAR for mobile robot

Abstract

In the real world, safe navigation is one of the primary aims of a mobile robot. For this to be effectively achieved, we must take into account both static and moving obstacles, including sidewalk curbs, ambulatory humans, and moving bicycles. To this end, we have developed a new emergency obstacle avoidance module for moving robots that uses Light Detection and Ranging (LIDAR) to detect static and moving obstacles. The module comprises a tilted LIDAR and a high speed embedded microcontroller. Tilting LIDAR enables the detection of lower sidewalk curbs, which is important for enabling safe navigation. The microcontroller is used for the estimation of obstacle motion and to facilitate rapid initiation of emergency obstacle avoidance maneuvers. By employing a fast embedded microcontroller, we can distinguish between lower-height obstacles and slopes, even on uneven surfaces, in real-time.