Motion Planning (CS 526)

2014 Fall
Faculty of Engineering and Natural Sciences
Computer Sci.& Eng.(CS)
3
10
Esra Erdem esraerdem@sabanciuniv.edu,
Click here to view.
English
Doctoral, Master
--
Formal lecture,Interactive lecture
Interactive,Project based learning,Guided discovery,Case Study
Click here to view.

CONTENT

The goal of motion planning is to enable robots to automatically compute their motions from high-level descriptions of tasks and models acquired through sensing. Over the years, motion planning has become a major research area in robotics. The techniques developed for robotics were not only used to create robots with motion autonomy, such as mobile robots navigating indoors, but also used in other domains, such as computer animation, computer-aided design, verification of building codes, exploration of virtual environments, and computational biology. Today, progress in motion planning is increasingly motivated by these applications. is to provide a coherent framework of motion planning algorithms. There will also be a discussion of existing methods to solve specific problems. Examples to the algorithms will be presented from the domains of mechanical design, manufacturing, medical surgery, computational biology. The focus will be on robust, efficient, and practical algorithms, with some form of provable guarantee of performance, over purely heuristic techniques or with optimum worst-case performance. self-contained; however the students are expected to have an interest in geometry and algorithms, and to have the skills to complete a significant programming assignment. The course will benefit students who may come from different backgrounds (e.g., computer science, mechatronics, electrical engineering, etc.). This course is primarily based on the course `Motion planning?, by Prof. Jean-Claude Latombe at Stanford University. No textbook is required. Here is a tentative syllabus: 1. Overview, 2. Path planning for a point robot, 3. Configuration space of a robot, 4. Probabilistic Roadmaps, 5. Collision detection and distance computation, 6. Sampling and connection strategies for probabilistic roadmaps, 7. Critycality based motion planning : target finding, 8. Coordination of multiple robots, 9. Humanoid and legged robots, 10. Mapping and inspecting environments, 11. Navigation through virtual environments, 12. Target tracking and virtual cameras, 13. Motion of crowds and flocks, 14. Surgical planning, 15. Motion of bio-molecules.

LEARNING OUTCOMES

  • At the end of this course, students are expected to have an elementary but algorithmically solid understanding of motion planning (including topics like configuration spaces, search-based motion planning, sampling-based motion planning), as well as an understanding of some of its applications in robotics

ASSESSMENT METHODS and CRITERIA

  Percentage (%)
Assignment 30
Individual Project 50
Presentation 20

RECOMENDED or REQUIRED READINGS

Textbook

"Principles of Robot Motion" by Howie Choset, Kevin M. Lynch, Seth Hutchinson, George A. Kantor, Wolfram Burgard, Lydia E. Kavraki and Sebastian Thrun.

"Planning Algorithms" by Steven M. Lavalle.