Kabushiki Kaisha Toshiba

Anthony Portelli

Adnan Aijaz

Marius Jurt

A method for controlling the position of an updating agent in a decentralised multi-agent system, the method comprising: identifying a first neighbouring agent within a communicative range of the updating agent; estimating a first distance to the first neighbouring agent and a first direction to the first neighbouring agent; determining a movement direction based on the first direction to the first neighbouring agent; and determining a movement magnitude based on an activation function, the first distance to the first neighbouring agent, and a desired reference distance. The activation function is configured such that the greater the difference between the first distance to the neighbouring agent and the desired reference difference, the larger the movement magnitude. The method further comprises moving the updating agent based on the movement direction and the movement magnitude.

Swarm control algorithm to maintain mesh connectivity while assessing and optimizing areal coverage in unknown complex environments

A method for optimizing coverage of a wireless network of a plurality of mobile robots in an environment in which each robot includes an optical sensor module, a microprocessor, and a wireless communication module. The method includes: receiving, by a first robot in the network, signals from a second robot in the network; determining, by the first robot based on the signals, that the first robot or second robot do not fulfil a network coverage condition; selecting, by the first robot, at least two landmarks in the environment; and performing, by the first robot, a movement based on an angle between the two landmarks with respect to the first robot, to improve the network coverage condition.

Growing community of inventors

Bristol, United Kingdom

Anthony Portelli