Apply Simultaneous Localization And Mapping (SLAM) to derive landmark positions and robot localization from measures made by the robot
In this project, we'll be localizing a robot in a 2D grid world. The basis for simultaneous localization and mapping (SLAM) is to gather information from a robot's sensors and motions over time, and then use information about measurements and motion to re-construct a map of the world.
In other words, using robot sensor measurements and movement, SLAM technique allows to create a map of an environment from only sensor and motion data gathered by a robot, over time. SLAM gives a way to track the location of a robot in the world in real-time and identify the locations of landmarks such as buildings, trees, rocks, and other world features. This is an active area of research in the fields of robotics and autonomous systems.
This project is the third project of UDACITY's Computer Vision nanodegree. The project is broken up into three notebooks and a couple of helper functions:
- Notebook 1 : Robot Moving and Sensing
- Notebook 2 : Omega and Xi, Constraints
- Notebook 3 : Landmark Detection and Tracking
Note that a measurement noise and a motion noise (representing the instruments and machine's precision gauge) are included in all sensor measurements and movement data. More noise will affect predictions adversely. The higher motion / measurement noise we have, the higher the difference between predicted and real value of poses of the robot. Therefore it makes sense to use 1-over-noise as the confidence or "strength" factor. The higher the noise, the less confident the measurement. Noise plays the role of the standard deviation sigma. More noise will flatten the gaussian probability distribution of the measure (around its exact value).
Download the repository and execute the notebooks.
- Ground truth of robot positions and landmark localizations
- Estimated landmark positions and robot positions. These are calculated using the SLAM approach from only motion and measurement data. SLAM gives us a way to both localize a robot and build up a map of its environment as a robot moves and senses in real-time.
- Representation of the "world" mapped by the robot using detection measures as it moves.
- X are the landmarks
- O is the robot
- Improved detection precision of a fixed landmark pattern with higher number of moves and measurements from the robot (20 measures, 100 measures, 300 measures) to map the environment.
