Visual victim detection and quadrotor-swarm coordination control in search and rescue environment

Gustavo A. Cardona, Juan Ramirez-Rugeles, Eduardo Mojica-Nava, Juan M. Calderon


We propose a distributed victim-detection algorithm through visual information on quadrotors using Convolutional Neuronal Networks (CNN) in a search and rescue en- vironment. Firstly, we describe the navigation algorithm, which allows quadrotors to avoid collisions with both, other quadrotors and obstacles in the environment. Sec- ondly, when at least one quadrotor detects a possible victim, this robot causes its clos- est neighbors to disconnect from the main swarm and form a new sub-swarm around the victim, which will validate if it is a victim or not. Thus, a formation control that permits quadrotors to acquire more information is performed based on the well-known rendezvous consensus algorithm. Finally, from the acquired images from each quadro- tor, we process them using CNN aimed to identify potential victims in the disaster area. Given the uncertainty and dissimilarity of the victim detection measurement among quadrotors’ cameras in the image processing, the Estimation Consensus (EC) and Max-Estimation Consensus (M-EC) algorithms are proposed focusing on agree- ing over the victim detection estimation. We illustrate that M-EC delivers better results than EC in scenarios with poor visibility and uncertainty as produced by fire, visual occlusion, and smoke. The algorithm proves that distributed fashion can obtain a more accurate result in decision-making on whether or not there is a victim, showing ro- bustness under uncertainties and wrong measurements in comparison when a single quadrotor tries to do it. The well-functioning of the algorithm was evaluated carrying out a simulation using a virtual simulated search and rescue environment in V-Rep while using its quadrotor model.


consensus; convolutional; neural-networks; quadrotors; swarm-navigation; victim-detection;

Total views : 0 times

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

ISSN 2088-8708, e-ISSN 2722-2578