Lecture 6: Navigation Flashcards
How is navigation achieved (the 2 approaches)?
Locomotion and path-planning
What are the 3 locomotion categories?
Mechanical locomotion
Biomimetic locomotion
Legged locomotion
What are the 4 key questions asked in navigation, along with the 4 features of navigation used to answer them?
Where am I going? Mission planning
What is the best way there? Path planning
Where have I been? Mapping
Where am I? Localisation
What is the difference between qualitative path planning and quantitative path planning?
Qualitative path planning is topological navigation that is feature-based. Quantitative path planning is metric navigation.
How are localisation and planning related?
The robot uses where they are to make decisions about future movement.
What is SLAM?
Simultaneous Localisation and Mapping
What 3 methods can be used to achieve feature-based localisation? Explain one of them.
Marcov localisation is used for global localisation without initial known pose.
Extended Kalman filters are used for local localisation.
Iconic localisation uses raw sensor readings to match actual observations to expected observations.
How does vSLAM differ from SLAM?
vSLam does not require feature extraction or LIDAR for navigation.
How does RatSLAM work?
It is based on computational models of the hippocampus. It integrates place cells, head direction cells, and grid cells to form a map.
What is the difference between place cells, head direction cells, and grid cells in rats?
Place cells fire when the rat is at a specific location in the environment
Head direction cells fire when the rat’s head is at specific orientations
Grid cells fire in a metrically regular way across the whole surface of a given environment.
What is an open source library that can be used for robot navigation?
OpenRatSLAM
