Abstract
Reducing Disagreement by trying to make the machine drive like a person does is, in the best of cases, far from easy. For a start, each person drives differently, even if they have the same goals and/or guidelines. Second, anyone’s way of driving is likely going to depend on experience, condition and state, as well as on the vehicle. Finally, most people can not explain to satisfaction the whys of their reactive decisions so that all specifics can be fitted by a function. Traditionally, data blocks that can not be explained by mathematical expressions can be approximated a posteriori by statistical tools, like splines. Indeed, it is possible to capture amild amount of data from a path, processing it with the correct methodology and come up with a set of equations that can do the trick, as proven in the field of field of Inverse Kinematics. These equations would be different each time a new device is employed and they require some analysis and practice to be fixed. Besides, this approach does not leave much space to adaptation and requires calibration of the system for each specific user and, in some cases, whenever the person changes his/her way of driving permanently (rehabilitation or degenerative process). Alternatively, the robot could try to observe how the person acts and mimic his/her way of doing things. Indeed, kids learn to talk before ever reading any grammar book and they probably do not solve any equation to start walking. If we could just ask the wheelchair to learn from its driver, the engineer’s life would become far easier. This approach is known as learning by imitation.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2012 Springer Berlin Heidelberg
About this chapter
Cite this chapter
Urdiales, C. (2012). If I Only Had a Brain. In: Collaborative Assistive Robot for Mobility Enhancement (CARMEN). Intelligent Systems Reference Library, vol 27. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-24902-0_5
Download citation
DOI: https://doi.org/10.1007/978-3-642-24902-0_5
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-24901-3
Online ISBN: 978-3-642-24902-0
eBook Packages: EngineeringEngineering (R0)