Simultaneous Localization and Mapping Quadcopter
Authors:Callie Chen, Scott Kline, Harsh Shah, Giuseppe Tonini, Steven Viska
Mentor:Subodh Bhandari, Professor of Aerospace Engineering, California State Polytechnic University, Pomona
Many scenarios, both military and civilian, necessitate the mapping and reconnaissance of possibly hazardous unknown indoor environments. Examples of such scenarios include indoor search and rescue missions after natural disasters and soldiers securing buildings with possible hostile forces. These missions exhibit high risk of human life loss and employing unmanned remote controlled robots has the potential of preventing unnecessary injuries or deaths. However, the lack of obstacle avoidance capability has limited the use of these vehicles, especially in the Unmanned Aerial Vehicles field. The purpose of this research is to develop a quadcopter capable of 3-D mapping enclosed areas and obstacle avoidance. The system will utilize Simultaneous Localization and Mapping (SLAM) as its mapping technique. A Microsoft Kinect camera, which uses only depth data, will be employed to recreate real-time detailed 3-D representations of indoors scenes. The Microsoft Kinect sensor is composed of an infrared projector, an infrared monochrome camera and a RGB camera. This sensor obtains the 3-D images and depth map calculations. The program utilized in this project is SCENECT, an open source software based on the laser scan software SCENE by FARO. A 3-D depth map is constructed by reflected infrared signals that the infrared monochrome camera receives. The signals are produced from the infrared projector. Chromatic and 3-D depth images are obtained simultaneously with the object’s features by using the RGB camera. Ultrasonic range finders will later be implemented into the system to detect obstacles in its path and prevent the quad copter from colliding into them. Obstacle avoidance will be tested by moving objects closer to the range finder and by observing the response of the motors. Once the minimum avoidance criterion is met and implementation of Microsoft Kinect Camera is complete, flight testing phase will commence.