Robotics Basics - Deliberative Paradigm

The deliberative paradigm focuses on thinking on advance and ahead on the future to create a big plan to follow. This implies looking ahead at the outcome of the possible actions and to search for the sequence that makes the robot achieve the desired goal, making the decision a search problem: looking in the representation of a problem for a desired state and storing the actions to achieve it. This search problem can be optimized, that is, choosing between actions for the best one according to some criteria.

**Sense –> Plan –> Action **

The first architectures of the deliberative paradigm showed to the robotics community two major problems that had to be considered on the development of robotics: the closed world assumption 20211210202959 and the frame problem 20211210204216.

Pure deliberative architectures are not frequently used, mainly because they require heavy computation and memory costs and they are very niche targeted 20211210200259.

Deliberative architectures

• STRIPS 20211215193324
• NHC 20211215202542
• NIST RCS 20211215202630

Notes References

20210514183815 INDEX - Robotics

20210711200423 Robotics Basics - Control Paradigms

20211210200259 The major drawbacks and problems of the deliberative paradigm are its costs and portability/modularity

20211210202959 The Closed World Assumption Problem

20211210204216 The frame problem (representation)

References

(Murphy 2000)

Murphy, Robin. 2000. Introduction to AI Robotics. Intelligent Robotics and Autonomous Agents. Cambridge, Mass: MIT Press.