Multihierarchical Graph Search
IEEE Transactions on Pattern Analysis and Machine Intelligence
A Robust Layered Control System For a Mobile Robot
A Robust Layered Control System For a Mobile Robot
Assistive navigation of a robotic wheelchair using a multihierarchical model of the environment
Integrated Computer-Aided Engineering
Improving efficiency in mobile robot task planning through world abstraction
IEEE Transactions on Robotics
Hi-index | 0.00 |
Except for pure reactive robots, that do not work with any explicit representation of their world [1], an intelligent robot must possess some symbolic representation of its environment in order to reason, plan (prediction), and perform efficiently (due to the intractable amount of subsymbolic information acquired from the real world). We have been working on that area during the last decade, in particular exploring the advantages of using abstraction and multiple abstraction for modeling the environment of a mobile robot. In this sense we have addressed the following main issues: – Automatic construction of the model. Assistive approaches (involving human operators) are possible [4] but limit the autonomy of the robot. – Automatic optimization (adaptation) of the model, for coping with the different situations that the robot may face during its operation without constructing an entirely new model each time from scratch. – Coherence between the symbols in the model and the real world. This must be addressed as a dynamic procedure since the real world changes continuously. –Efficiency in using the model. We believe that the best model for a given robot is the one that improves the most the robot's planning of operations. Up to now, we have been working on obtaining a comprehensive solution with abstraction as a basis for coping with all these issues at once. In the next we describe in more detail our solutions to each one.