학술논문
카메라와 초음파 센서를 이용한 이동로봇의 원 궤적 경로 계획 기법 / Circular Path Planning Method for a Mobile Robot using Camera and Ultrasonic Sensor
Document Type
Dissertation/ Thesis
Author
Source
Subject
Language
Korean
Abstract
본 논문에서는 카메라와 초음파 센서를 사용하여 주변 환경에 대한 정보를 얻고 이동로봇이 목적지까지 안전하게 도착할 수 있도록 하기 위한 장애물 회피 알고리즘을 제안하였다.이동로봇의 전면에 존재하는 장애물 정보는 카메라를 통해서 얻어지는데, 신속한 장애물 정보 획득을 위하여 영상 축소 기법을 적용하였다. 카메라를 통해서 인지할 수 없는 장애물이나 사각지역에 존재하는 장애물 정보를 획득하기 위해서 카메라와 초음파 센서의 데이터를 융합하였다. 데이터는 격자 지도를 통하여 융합되었는데, 격자 지도는 이동로봇이 주행하는 동안 획득하는 모든 데이터를 저장함으로써 보다 효율적인 경로 생성을 가능하게 하였다. 장애물 회피 경로를 생성하는 방법으로는 원 궤적 경로 계획 알고리즘을 제안하였는데, 다수의 장애물이 존재하는 상황에서도 효율적인 이동 경로를 생성하였다.생성된 경로를 따라 이동하기 위해서 이동로봇의 주행 속도를 결정해야 하는데, 이 때 주행 중에 발생하는 원심력을 고려함으로써 보다 안정된 주행을 확보하였다. 본 논문에서 제안한 장애물 회피 알고리즘의 성능을 입증하기 위하여 PIONEER2-DX 이동로봇에 적용한 실험을 수행하였다.
This thesis proposes an obstacle avoidance algorithm for a mobile robot. Using a camera and ultrasonic sensors, the mobile robot obtains the circumferential information around the mobile robot. The achieved information helps the robot fast and safely navigating to the goal point.The mobile robot obtains the information about the obstacles lied in the front through a camera. An image reducing method is proposed to improve the processing speed. For the obstacle’s information which cannot be acquired through a camera or an obstacle lied on the dead zone of camera, a data fusion method is proposed in which the data of camera and the data of ultrasonic sensors are combined together. The data is fused through a grid map which helps to generate more effective path finding by saving the data which the robot is acquired through navigating to the goal point. As an obstacle avoidance method, this thesis proposes a circular path planning algorithm. This proposed algorithm generates an ideal moving path for a mobile robot in the environment where there are many obstacles. By deciding the robot’s moving speed to have the robot navigate following the generated path, we can successfully secure the driving stability by considering the centrifugal force generated during the navigation. To demonstrate the efficiency of the proposed obstacle avoidance algorithm, we performed several experiments using a Pioneer-2dx mobile robot.
This thesis proposes an obstacle avoidance algorithm for a mobile robot. Using a camera and ultrasonic sensors, the mobile robot obtains the circumferential information around the mobile robot. The achieved information helps the robot fast and safely navigating to the goal point.The mobile robot obtains the information about the obstacles lied in the front through a camera. An image reducing method is proposed to improve the processing speed. For the obstacle’s information which cannot be acquired through a camera or an obstacle lied on the dead zone of camera, a data fusion method is proposed in which the data of camera and the data of ultrasonic sensors are combined together. The data is fused through a grid map which helps to generate more effective path finding by saving the data which the robot is acquired through navigating to the goal point. As an obstacle avoidance method, this thesis proposes a circular path planning algorithm. This proposed algorithm generates an ideal moving path for a mobile robot in the environment where there are many obstacles. By deciding the robot’s moving speed to have the robot navigate following the generated path, we can successfully secure the driving stability by considering the centrifugal force generated during the navigation. To demonstrate the efficiency of the proposed obstacle avoidance algorithm, we performed several experiments using a Pioneer-2dx mobile robot.