Abstract
This work addresses the problem of traction control in mobile wheeled robots in the particular case of the RoboCup Middle Size League (MSL). The slip control problem is formulated using simple friction models for ISePorto Team Robots with a differential wheel configuration. Traction was also characterized experimentally in the MSL scenario for relevant game events. This work proposes a hierarchical traction control architecture which relies on local slip detection and control at each wheel, with relevant information being relayed to a higher level responsible for global robot motion control. A dedicated one axis control embedded hardware subsystem allowing complex local control, high frequency current sensing and odometric information procession was developed. This local axis control board is integrated in a distributed system using CAN bus communications. The slipping observer was implemented in the axis control hardware nodes integrated in the ISePorto Robots and was used to control and detect loss of traction. An external vision system was used to perform a qualitative analysis of the slip detection and observer performance results are presented.
Highlights
Traction control is becoming a relevant problem in RoboCup Middle Size League (MSL) competitions
José Almeida, André Dias, Alfredo Martins, João Sequeira and Eduardo Silva: 5 Distributed Active Traction Control System Applied to the RoboCup Middle Size League
The problem of traction in mobile wheeled robots in the RoboCup MSL league scenario was analysed in this paper
Summary
Traction control is becoming a relevant problem in RoboCup Middle Size League (MSL) competitions. To the majority of the authors in the literature the objective of traction is to reduce the slip rate between the surface and the wheels This makes it possible to minimize power consumption, improve motion control performance (trajectory tracking) and overall vehicle safety (reducing the vehicle’s effective braking distance). In a similar approach to bio-inspired control, this work is part of a larger framework where this prio-perceptive motion control plays a relevant role in the application of mobile robotics in extreme environments and for high motion performance situations To support this approach, and contrary to the approaches mentioned above, the solution provided does not require prior knowledge and modelling of the wheel-ground interaction. The paper concludes with a short discussion on the results obtained and outlines topics that may be studied in the future
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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 call
