The Multi-degree-of-freedom (MDOF) Mobile Robot with Compliant Linkage
by Johann Borenstein, The University of Michigan
ABSTRACT
Multi-degree-of-freedom (MDOF) vehicles have many potential advantages over conventional (i.e., 2-DOF) vehicles. For example, MDOF vehicles can travel sideways and they can negotiate tight turns more easily. In addition, some MDOF designs provide better payload capability, better traction, and improved static and dynamic stability. However, MDOF vehicles with more than three degrees-of-freedom are difficult to control because of their overconstrained nature. These difficulties translate into severe wheel slippage or jerky motion under certain driving conditions. These problems make it difficult to benefit from the special motion capabilities of MDOF vehicles in autonomous or semi-autonomous mobile robot applications. This is so because mobile robots (unlike automated guided vehicles) usually rely on dead-reckoning between periodic absolute position updates and their performance is diminished by excessive wheel slippage.
This paper introduces compliant linkage, a new concept in the control and kinematic design of MDOF mobile robots. As the name implies, compliant linkage provides compliance between the drive wheels or drive axles of a vehicle, to accommodate control errors which would otherwise cause wheel slippage.
The concept of compliant linkage was implemented and tested on a 4-DOF vehicle built at the University of Michigan's Mobile Robotics Lab. Experimental results show that control errors are effectively absorbed by the linkage, resulting in smooth and precise motion. The dead-reckoning accuracy of the compliant linkage vehicle is substantially better than that reported in the literature for other MDOF vehicles; it was found equal to, or even better than that of comparable 2-DOF vehicles.
Figure 1 shows our MDOF vehicle with compliant linkage. The vehicle comprises of two trucks (in our prototype, these are commercially available LabMate robots from TRC. The two trucks are connected by the compliant linkage, which allows almost force-free relative motion within its physical range. A linear encoder measures the momentary distance between the two trucks, and two absolute rotary encoders measure the rotation of the trucks relative to the compliant linkage. Each of the four drive wheels in the system has a shaft encoder to allow conventional dead reckoning.
Project History
The MDOF vehicle project started in 10/91 and will end (after a 1-year no-cost extension) in 9/95. Johann Borenstein is the PI in this project, which was originally funded by an NSF grant over $229,500. Later, the Department of Energy added partial support to the project by funding additional research work. The project has been rather successful: During all stages of the project work progress was ahead of schedule. The final prototype was successfully demonstrated one year ahead of schedule. In the remaining time the project went far beyond the original workscope: We developed and implemented a unique and innovative method for correcting dead-reckoning errors in mobile robots (called
"IPEC"). In the course of the project we also studied errors in mobile robot dead-reckoning and developed methods for measuring and reducing such errors (see work on the UMBmark procedure).
The many results of this project were disseminated in 14 publications (3 Journal papers, 9 conference papers, 2 video proceedings (IEEE R&A). Two patents have been issued.
In 1996 the first commercial prototype of the MDOF vehicle, called "OmniMate," was built and tested.
For more details on the MDOF vehicle with compliant linkage see papers 34, 35, 36, 47, video 01
For more details on dead-reckoning error correction with IPEC see papers 48, 49, 53, 57 video 02
For more details on measuring and corrrecting systematic dead-reckoning error with UMBmark see papers 8,59,60,61
This file last updated on 7/4/96 by Johann Borenstein
johannb@umich.edu