Abstract
Dams are important engineering facilities in the water conservancy industry. They have many functions, such as flood control, electric power generation, irrigation, water supply, shipping, etc. Therefore, their long-term safety is crucial to operational stability. Because of the complexity of the dam environment, robots with various kinds of sensors are a good choice to replace humans to perform a surveillance job. In this paper, an autonomous system design is proposed for dam ground surveillance robots, which includes general solution, electromechanical layout, sensors scheme, and navigation method. A strong and agile skid-steered mobile robot body platform is designed and created, which can be controlled accurately based on an MCU and an onboard IMU. A novel low-cost LiDAR is adopted for odometry estimation. To realize more robust localization results, two Kalman filter loops are used with the robot kinematic model to fuse wheel encoder, IMU, LiDAR odometry, and a low-cost GNSS receiver data. Besides, a recognition network based on YOLO v3 is deployed to realize real-time recognition of cracks and people during surveillance. As a system, by connecting the robot, the cloud server and the users with IOT technology, the proposed solution could be more robust and practical.
Highlights
In the water conservancy industry, there are many fundamental engineering facilities, such as dams, water and soil conservation, water transfer project, shipping project, water supply project, hydraulic power plants, irrigation facilities, etc
This paper presents a general system for the mobile robot, which includes the robot, the cloud server, and terminals
The method here avoids high computing burden since we only use the data for odometry calculation, not for real-time mapping
Summary
In the water conservancy industry, there are many fundamental engineering facilities, such as dams, water and soil conservation, water transfer project, shipping project, water supply project, hydraulic power plants, irrigation facilities, etc. As the robot body itself, we design a small powerful wheeled skid-steering mobile platform. Compared to alternative wheel configurations such as Ackerman or axle-articulated, the skid-steering platform shows two major advantages It is simple and robust in mechanical terms. The method here avoids high computing burden since we only use the data for odometry calculation, not for real-time mapping. The first EKF node fuses IMU data, wheel odometry, LiDAR odometry to get robot transform in the inertial frame. By running it on the robot STM32 MCU board, we can ensure the fusion has real-time performance with high frequency.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
