As a control algorithm moves robots through a space, each state the robots can be in is represented by a spot on the 3D graph. As the algorithm explores new options, it finds failures and potential next moves. This graph represents all of the wrong moves explored (blue) and the one route that is correct (red).
A new approach to designing motion plans for multiple robots grows “trees” in the search space to solve complex problems in a fraction of the time
In a building several stories tall with numerous rooms, hundreds of obstacles and thousands of places to inspect, the several dozen robots move as one cohesive unit. They spread out in a search pattern to thoroughly check the entire building while simultaneously splitting tasks so as to not waste time doubling back on their own paths or re-checking places other robots have already visited.
Such cohesion would be difficult for human controllers to achieve, let alone for an artificial controller to compute in real-time.
“If a control problem has three or four robots that live in a world with only a handful of rooms, and if the collaborative task is specified by simple logic rules, there are state-of-the-art tools that can compute an optimal solution that satisfies the task in a reasonable amount of time,” said Michael M. Zavlanos, the Mary Milus Yoh and Harold L. Yoh, Jr. Associate Professor of Mechanical Engineering and Materials Science at Duke University.
“And if you don’t care about the best solution possible, you can solve for a few more rooms and more complex tasks in a matter of minutes, but still only a dozen robots tops,” Zavlanos said. “Any more than that, and current algorithms are unable to overcome the sheer volume of possibilities in finding a solution.”
The researchers show that this method will always find an answer if there is one, and it will always eventually find the best one possible. They also show that this method can arrive at that answer exponentially fast. Working with a problem of 10 robots searching through a 50-by-50 grid space— 250 houses to pick up mail — current state-of-the-art algorithms take 30 minutes to find an optimal solution.
STyLuS* does it in about 20 seconds.
via Duke University
The Latest Updates from Bing News & Google News
Go deeper with Bing News on:
- Stare at the Abyss; It Might Look Back
Right, Jimmy?” you look at the robot you’re acquaintances with, not wanting more maniac animatronics trying to squeeze your head like an orange. “We’re all friends here, okay?” Half a minute passes in ...
- UAE warns of swarming drone assassins in ‘dystopian future’ war
UAE ambassador to the UN Lana Nusseibeh warned on Monday of swarms of assassin drones being sent out in cross-border raids, as groups such as Yemen’s Houthi rebels incorporate ever-deadlier smart ...
- Love, Death and Robots: Vol. 3 Review
Love, Death and Robots Vol. 3 is available now on Netflix. Netflix’s ongoing animation anthology Love, Death and Robots returns for a third volume of brand-new shorts with the unofficial mandate of ...
- Who hacked the slaughterhouse? When robots and AI take over farms
Advanced tech in farming promises to maximize crop yields, minimize the impact of climate change, and fight global food insecurity. Yet swift adoption without consideration of the risks may increase ...
- Love, Death and Robots season 4 remains TBA, but another volume is a no-brainer
Whilst Love, Death and Robots season 4 remains unconfirmed, the hit animated anthology is highly likely to return for another volume on Netflix, here’s why.
Go deeper with Google Headlines on:
Go deeper with Bing News on:
Coordinating movements of multiple robots
- Michael Pavitt: Farewell as my time at insidethegames draws to a close
This week marked my final one at insidethegames, as I hang up the green and pink tie after close to seven-and-a-half years.
- DeepMind's 'Gato' is mediocre, so why did they build it?
DeepMind's program is a generalist, to test the notion that over time, greater computing power will win in AI.
- Wrangling RC Servos Becoming A Hassle? Try Serial Bus Servos!
Coordinated Action: We could only send commands to one servo at a time, but sometimes it’s important for multiple servos ... allowing much smoother robot movements. The DRS-0101 is also the ...
- Qualcomm plunges into the robotics market with new platform
One of the current constraints of real-world autonomy is that on-board data-processing power is expensive -- and coordination across multiple autonomous ... years of slow movement, the FAA is ...
- NeuroMechFly: A digital twin of Drosophila
The twin is a big step towards reverse engineering the neuromechanical control of animal behavior, and developing bioinspired robots ... uses multiple camera views to quantify the movements ...