Geometric Recognition of Diverse Terrain in Real-Time for a Six-Wheeled Robot based on Laser Scanning Sensors
Nabih Pico, Eui-Chan Kim, Sang-Hyeon Park, Meseret Tadese, Huy Nguyen Tran, Beomjoon Lee, Hyungpil Moon
- Year
- 2022
- Citations
- 4
Abstract
Wheeled robots are involved in different applications due to their adaptability to different terrains. The stability of the robot is related to the permanent contact of the wheels with the ground. This paper presents the method to recognize the geometry terrains in real-time when the wheel has contact at one or multiple contact points with the soil by using laser scanning sensors and measuring the contact angle between the wheel-ground. Furthermore, the robot can recognize when its wheel loses contact with the terrain. Thus, we can create a proper robot control by obtaining the Jacobian matrix and its inverse related to the terrain information. In consequence, the robot can overcome diverse terrains and move safely. The method is validated in experimental results when the six-wheeled robot recognizes a rocky terrain and a slope with a step.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Fractional Differential Equations
Igor Podlubný
2025
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991