Publish at October 20 2020 Updated February 14 2024
Date: November 10 2020 to November 10 2020
From assisted wheelchairs and delivery robots to robotic assistants in hospitals and nursing homes, robots are frequently called upon to operate and circulate in unprotected public environments, such as busy streets and crowds. The field of cobotics - working with robots - is opening up, and the challenges posed extend beyond the technical realm into that of immense human complexity.
For example, legislators and regulators need guidelines to define what is acceptable in our relations with robots, and what should be demanded of them. Programmers need to know exactly how humans react and behave in various situations, such as when passing through a doorway, when two streams of people cross, when a sudden obstacle (a person who stops) presents itself in a stream of people, and so on.
Crowdbot, a consortium funded by the European ICT H2020 program and coordinated by Inria, is addressing these issues. In a huge room, they are analyzing crowd movements in various situations and, from there, modeling these movements to then make a virtual robot evolve in a virtual crowd in order to speed up the development process at lower cost and without the risk of causing accidents.
This video shows data collection
Algorithms can then be developed and quickly compared in the simulator. In this way, solutions can be optimized more quickly and then tested in situ.
For the full article: Simulating robot navigation in crowds
Domaine de Voluceau
Rocquencourt - B.P. 105
78153 Le Chesnay
France
Tél.: 33 (0)1 39 63 55 11
You might also like:
Non-periodic acoustic insulators for low frequencies
"We don't know how far we'll be able to go. But the important thing is to bring out original, usable vibro-acoustic phenomena.
Present security compromised by the future: at the dawn of quantum computers
A race is on between the developers of quantum computers and cryptographers who seek to ensure data confidentiality because the most advanced current protocols in cryptography are challenged by the capabilities of quantum computers.
Ensuring the transparency of automated administrative decisions
How can we ensure that administrative algorithms apply the law as interpreted by administrative legal departments, without distortion or approximation?
Identifying biodiversity on a large scale
How can you identify a 10 km2 profusion of leaves and tangled plants?
Graph-based learning for frugal artificial intelligence
Data modeled in graphs (which can be multi-dimensional) allow the detection of relationships that are practically invisible in other forms and, above all, with much less computational and energy resources.
