Long-Thin Multisensory Mobile Robot For Interior Debris Searching

The earthquake and tsunami accidents always happen frequently in the world and the buildings would be broken down. Interior debris searching survivers is a very urgent but also complex and difficult task. In order to improve the efficiency of the search and rescue, and protect the safety of the rescuers, the search and rescue robots are developed strongly domestic and overseas. The debris environment is disorderly and narrow, so there are many challenges for robot mobility, localizability and human detection technology. The developed rescue robots used for debris surface environment are large, commonly. The robot locomotivity is weak in the narrow interior debris environment and rugged ground surface, so robots can't deep inside the debris. Because of the invisibility of closed debris environment, the rescuers can't judge the pose of the robots visually. The 3d localization of robot in the debris environment is difficulty for rescue robot technology. In order to detect human and navigation, the search and rescue robots are usually equipped with camera and other multi-sensor system. But these robots provide limited sensor function, can't get temperature, CO₂ concentration, audio and other sensor information. The volume of sensors is large, commonly. The communication of sensors data is not integrated and the transmitted distance is not very long.In view of the search and rescue mission inside the debris environment, this article developed a long thin multisensory mobile robot. The robot body is composed of active segment and passive segment. Locomotion driving and head orientation adjustment is executed by active segment. Passive segment uses Flexible Sensor Tube (FST) to realize 3D pose measurement of the robot body and obtain the 3D localization of survivors consequently. Standardised mechanical and electrical interfaces facilitate the multi-segment expansion of robot body for long distance searching. A front multi-sensor unit is mounted to the head of robot for video capture, temperature and CO₂ concentration data collection, and two-way audio conversation between rescuers and survivors. Highly integration of embedded system and micro sensors minimizes the system and improves the ability of coming through narrow environment. The main content of this paper is summarized as bellow.Firstly, the new robot body structure is developed, which is composed of the passive segment and the active segment. The passive segment has many joints and the active segment is composed of power unit and driving unit.The encoders are installed inside the passive segment, which read the rotation angle of joints. The active segment has one motion control module that is composed of a conctroller, a DC driving module and 3 DC motors. The control module drives Locomotion and head orientation adjustment of the robot.Secondly, a front embedded multi-sensor unit is mounted to the head of robot. This unit is capable of collecting video, audio, temperature, CO₂, and other sensor information. It can also complete two-way audio conversation and provide LED lights for camera.Thirdly, the kinematic of robot is derived. After reading the joints, the robot localizes the head comparing to the robot base. 3d graph of robot is built in the monitoring terminal basing on OpenGL, so the rescuers can visualize terrain conditions of the debris environment. With the help of the inverse kinematics of active segment and joint motor closed-loop control algorithm, head orientation adjustment of the robot can be controlled.Fourthly, human-robot interface software of terminal is written to compelete 3d modeling and positioning, robot motion control and monitoring of front multi-sensor unit.Lastly, the experiments of real robot prototype conducted in a simulated debris environment demonstrate the search and rescue performance of the robot system and verified the ability to move in debris environment, 3d positioning in closed space, and searching and detection of muti-sensor information of this robot.The robot can be used for debris environment searching and survivors rescue, which improves the efficiency of rescue, and protects the rescuers.