AI News, Robots working as a group are able to determine the optimal order of their tasks

Robots working as a group are able to determine the optimal order of their tasks

The researchers from the IRIDIA laboratory have based their study on swarm robotics, a branch of robotics that draws from the collective and organised behaviour of social animals (such as ants) in order to create groups of robots that exhibit artificial intelligence.

In order to solve this problem, some of the robots gradually form a chain between the three points in space, which the others use as a guide as they test the various possible combinations by following instructions from the robots who make up the chain (see photo).

This ability to plan ahead is considered to be a complex cognitive skill, and it emerges from the interactions between the individuals in the group: together, the robots are able to plan a sequence of actions, which no individual in the group would be able to do alone.

Working Together, These Robots Solved a Problem

The researchers based their work on swarm robotics, a branch of robotics that draws from the collective and organized behavior of social animals (such as ants) in order to create groups of robots that exhibit artificial intelligence.

For their latest research, Mauro Birattari and Lorenzo Garattoni created a swarm of robots able to perform a sequence of three actions, without knowing the correct order in advance.

The research team says that the study demonstrates that robots are able to collectively determine a sequence of actions whose required order was previously unknown.

The possibilities that the researchers anticipate include searching for survivors after a natural disaster, exploring unknown or hostile environments, building structures on dangerous sites and various applications in agriculture.

In 2016, researchers at North Carolina State University developed a combination of software and hardware to allow the use of unmanned aerial vehicles and insect cyborgs, or biobots, to map large, unfamiliar areas —

Robot See, Robot Do: How Robots Can Learn New Tasks by Observing

It can take weeks to reprogram an industrial robot to perform a complicated new task, which makes retooling a modern manufacturing line painfully expensive and slow.

“We ask an expert to show the robot a task, and let the robot figure out most parts of sequences of things it needs to do, and then fine-tune things to make it work.” At a recent conference in St. Louis, the researchers demonstrated a cocktail-making robot that uses the approaches they’re working on.

The robot—a two-armed industrial machine made by a Boston-based company called Rethink Robotics, watched a person mix a drink by pouring liquid from several bottles into a jug, and would then copy those actions, grasping bottles in the correct order before pouring the right quantities into the jug.

Watching thousands of YouTube videos may sound time-consuming, but the learning approach is more efficient than programming a robot to handle countless different items, and it can enable a robot to deal with a new object.

The learning systems used for the grasping work involved advanced artificial neural networks, which have seen rapid progress in recent years and are now being used in many areas of robotics.

Recipes for Robots

And although most robot grippers (hands) are designed to pick up a single, predefined object, those in von Wichert’s lab can use the same gripper to grasp a wide variety of objects – without hardware alterations.  “What’s new about what we have developed is the way in which our system takes a generalized task description and automatically composes its detailed behavior from that,” says Wurm

RI Seminar: Stefan Schaal : From Movement Primitives to Associative Skill Memories

Stefan Schaal Professor of Computer Science, Neuroscience, and Biomedical Engineering, University of Southern California March 28, 2014 Abstract ...

8 ADVANCED ROBOTS ANIMAL YOU NEED TO SEE

8 Advance Robots Animal You NEED To See 10 Amazing Robots That Really Exist 5 Advanced Robots Animal You NEED To See 1.BionicANTs | Festo ...

Fast Target Prediction of Human Reaching Motion for Cooperative Human-Robot Manipulation Tasks

Claudia Pérez-D'Arpino, Julie A. Shah. Fast Target Prediction of Human Reaching Motion for Cooperative Human-Robot Manipulation Tasks Using Time Series ...

Physical Symbol Grounding and Instance Learning through Demonstration and Eye Tracking

It is natural for humans to work with abstract plans which are often an intuitive and concise way to represent a task. However, high level task descriptions contain ...

Life in the Lab: A DNA sequencing pipeline

Louise Aigrain is a senior staff scientist, working in DNA Pipelines Research and Development group at the Wellcome Trust Sanger Institute. In this film she ...

Humanoidly Speaking - Learning about the world and language with a humanoid friendly robot

Accepted video of the International Joint Conference on Artificial Neural Networks, Video Competition, 2015. Authors: Xavier Hinaut, Johannes Twiefel, Marcelo ...

Human-Robot Co-manipulation: a robotic application using iTaSC methodology

This video shows an application of the iTaSC-methodology to an overconstrained task involving human–robot co-manipulation A robot assists an operator to ...

A Curious Robot Learner for Interactive Goal-Babbling

An updated version of the video is available at ------ My PhD Thesis : The challenges .

Robin Hanson: "The Age of Em: Work, Love and Life when Robots Rule the Earth" | Talks at Google

Robin Hanson, research associate at the Future of Humanity Institute of Oxford University, speaks to his upcoming book, "The Age of Em: Work, Love, and Life ...

Le Lab presents The Mechanical Side of Artificial Intelligence with Robert Wood

Wyss Core Faculty member, Robert Wood, Ph.D., presents a talk titled “The Mechanical Side of Artificial Intelligence”. Artificial intelligence typically focuses on ...