Workshop (sadna)
Fall 2023
0368-3535-01
Course hours: Wednesday, 14:00-16:00
Location: Check Point 180
Instructor: Dan Halperin, [email protected]
Office hours by appointment
Teaching assistant: Dror Livnat ([email protected])
Software helpdesk: Michael Bilevich (michaelmoshe at mail tau ac il)
Projects: Motion planning algorithms compute collision-free motion paths for objects that move among obstacles. They arise in robotics, graphical animation, surgical planning, navigation systems, computational biology and computer games, among other domains. In this workshop we will explore and apply algorithms for one of several robot systems. For example: (i) an industrial collaborative robotic arm (UR5e) playing Jenga or Rush-hour, or even writing on the board and erasing it, etc. (ii) a desk top robotic arm (myCobot 320) using chopsticks (iii) an autonomous vehicle (RoboMaster) mapping and/or navigating the floor (iv) drones that need to carry out a prescribed task while avoiding obstacles on their way as well as other drones.
A typical project involves implementing motion planning algorithms for a robotic arm, a drone, or a mobile robot in the real world.
Class meetings
At the beginning of the workshop, the very basics of motion planning algorithms will be presented, and possible projects will be laid out. Below is the complete list of expected class meetings.
- 18/10/23 – Introduction to robot motion planning + suggested projects
- 25/10/23 – More background, review of software tools
- 15/11/23 – Presentation of project plan
- 20/12/23 – Presentation of POC
- TBD – Presentation of the final project
The project is intended to involve a significant amount of teamwork. The recommended team size is three. Each team will design a project together with the course staff, develop it along the semester and the exams’ break, and submit it at a TBD date.
Bibliography: click here
Projects: Following are a few past projects.
Planning the motion of the Kinova arm to translate lesson plans into drawing on a board
Bouncing a balloon using two drowns that play in turns, keeping the balloon in the air
Estimating the location of a Robomaster robotic vehicle based on few distance measurements in a know environment
Passing cargo in a known environment with obstacles as fast as possible without dropping it, using Robomaster robotic vehicle
Sorting different types of waste using multi-arm UR5 setting in PyBullet simulation