Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Spin shots for a robotic billiard player
This project aims to take advanced shots using cueball spin with a robotic arm that has been developed to play billiards. It is an exciting opportunity to develop the vision-based localisation and shot-taking control needed to take such shots.
To start in September 2019 or earlier.
The Automatic Control Lab in D-ITET is building a robot snooker player, with the ultimate aim of challenging the world champion. Currently several students are working on the vision system, artificial intelligence unit, and cue control. The next step is to extend the range of shots to include cueball spin.
Snooker is much more challenging than other billiard games owing to the very large table (1.8 x 3.8m) and tight pockets relative to the ball size. Long shots are difficult even for professionals. In this project we will use a robot arm, with a "vision-in-the-loop" feedback system based on a cue-mounted camera, to localise the cue with respect to the balls and trigger spin shots.
The Automatic Control Lab in D-ITET is building a robot snooker player, with the ultimate aim of challenging the world champion. Currently several students are working on the vision system, artificial intelligence unit, and cue control. The next step is to extend the range of shots to include cueball spin.
Snooker is much more challenging than other billiard games owing to the very large table (1.8 x 3.8m) and tight pockets relative to the ball size. Long shots are difficult even for professionals. In this project we will use a robot arm, with a "vision-in-the-loop" feedback system based on a cue-mounted camera, to localise the cue with respect to the balls and trigger spin shots.
The project contains several steps:
1) Review literature on real-time localization techniques and the previous student project that was carried out on this topic.
2) Extend the mathematical model of the vision task to one in which the cue is not constrained to strike the cue ball from a fixed height and angle. The vision task will make use the geometry of the billiard table, balls, and optionally the table markings.
3) Make appropriate modifications to the filter that reconciles the images seen by the overhead ceiling camera and cue-mounted camera.
4) Test the cueing performance by measuring and logging shots of different kinds.
The project contains several steps:
1) Review literature on real-time localization techniques and the previous student project that was carried out on this topic.
2) Extend the mathematical model of the vision task to one in which the cue is not constrained to strike the cue ball from a fixed height and angle. The vision task will make use the geometry of the billiard table, balls, and optionally the table markings.
3) Make appropriate modifications to the filter that reconciles the images seen by the overhead ceiling camera and cue-mounted camera.
4) Test the cueing performance by measuring and logging shots of different kinds.
Joe Warrington and Nikos Kariotoglou, Automatic Control Lab (warrington@control.ee.ethz.ch, karioto@control.ee.ethz.ch)
Joe Warrington and Nikos Kariotoglou, Automatic Control Lab (warrington@control.ee.ethz.ch, karioto@control.ee.ethz.ch)