Swarm Optimized Cartesian Ping-Pong, Anyone?

Posted 16 Nov 2012 at 20:37 UTC by steve Share This

Engineers love to do crazy things and when they involve robots, we love to tell you about them. We've reported on a lot of ping-pong playing robots over the years but usually they're based on conventional industrial robot arms or humanoid arm designs. What if, instead of a multi-jointed arm, you wanted to design a cartesian ping-pong playing robot? That is, a robot that can only move linearly on an X, Y, and Z axis. That's the question Hossein Jahandideh and his fellow engineers asked themselves. And to make things more interesting, they used a Particle Swarm Optimization (PSO) algorithm to plan for the approach of the incoming ping-pong balls. The result is the paper, "Ball Striking Algorithm for a 3 DOF Ping-Pong Playing Robot Based on Particle Swarm Optimziation" (PDF format). From the paper:

"It has been shown that a robot as simple and low cost as a Cartesian robot holding a standard racket can be programmed to play ping-pong against a human player. A PSO-based algorithm was proposed to determine when and how to hit the ball. This algorithm, aside from having a near perfect success rate at throwing the ball to a specified target, can also be adjusted to follow various strategies, such as the ball reaching the target with maximum speed, or with maximum spin, etc."

The paper covers all the math developed to build and test a simulation of the cartesian ping-pong playing robot. The construction of an actual robot is still in the planning stages. We look forward to seeing video of their results. The authors note that the algorithms developed in the paper should be applicable to conventional, non-cartesian robots as well.

See more of the latest robot news!

Recent blogs

25 Feb 2017 steve (Master)
19 Feb 2017 mwaibel (Master)
16 Feb 2017 AI4U (Observer)
17 Jan 2017 shimniok (Journeyer)
16 Aug 2016 Flanneltron (Journeyer)
9 Jul 2016 evilrobots (Observer)
27 Jun 2016 Petar.Kormushev (Master)
2 May 2016 motters (Master)
6 Nov 2015 wedesoft (Master)
10 Sep 2015 svo (Master)
Share this page