Lecture 6

Advanced Sensing: Place Recognition and Occupancy Mapping
of 18
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Robotics Lecture 6: Advanced Sensing: Place Recognitionand Occupancy Mapping See course website  for up todate information. Andrew DavisonDepartment of ComputingImperial College London  Review: Monte Carlo Localisation Practical 5 OA BC DE FG xy θ abcdefgh ã  Most challenging part is getting the likelihood function right tocorrectly reweight particles after measurements.  Global Localisation (‘Kidnapped Robot’) with Sonar ã  In MCL, global localisation can be attempted by initialising a largenumber of particles randomly spread through the environment, andthen running the filter normally. However, this requires manyparticles (computationally expensive) and it may take manymovements and measurements to find the right location. ã  We would expect better performance with more informative sensing:e.g. a ring of sonar sensors all making measurements at the sametime rather than just one.  One Depth Measurement and Resampling ã  After a measurement (e.g. sonar depth = 20cm), the weights of particles consistent with it will increase. ã  Movement and further measurements are needed to lock downposition, and ambiguities may still arise.
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks