Dempster-shafter method for localization and map making

Dempster-shafter localization is similar to bayesian localization 20220301203202. It uses instead dempster-shafter thoery 20220302200105 for deciding if a cell is either empty, occupied or dontknow.

Sonar belief function

Based on the sonar sensor model 20220228203000, the belief function of a sonar reading is:

For Region I:

\(m(Occupied)=\\frac{(\\frac{R-r}{R})+(\\frac{\\beta-\\alpha}{\\beta})}{2} \* Max\_{occupied}\) m(Empty) = 0.0 m(dontknow*) = 1 − *m*(*Occupied) For Region II:

m(Occupied*) = 0.0 \(m(Empty) = \\frac{(\\frac{R-r}{R})+(\\frac{\\beta-\\alpha}{\\beta})}{2}\) *m*(*dontknow) = 1 − m(Empty)

Notes References

20220228202048 Iconic localization and map making

20210514183815 INDEX - Robotics

References

(Murphy 2000)

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