This class will give an introduction to robotics. It will be structured into the following parts:
- Generating motion and and dynamic control: This chapter will cover coordinate frames, non-holonomic constraints, Ackermann-drive (in analogy to street cars), PID.
- Planning: Planning around obstacles, path finding, Dijkstra, A*, configuration space obstacles, RRTs, lattice planners, gradient methods, potential fields, splines.
- Localization and mapping: state estimation problem, Bayesian filter, Odometry, Particle & Kalman filter, Extended and Unscented Kalman-Filter, simultaneous localization and mapping (SLAM).
- Vision and perception: SIFT, HOG-features, Deformable parts models, hough transform, lane detection, 3d-point clouds, RANSAC .
After these lectures, students will be able to design basic algorithms for motion, control and state estimation for robotics.