Web-based configuration tool for benchmarking of simulated intervention autonomous underwater vehicles

Benchmarking is nowadays an issue on robotic research platforms, due to the fact that it is not easy to reproduce previous experiments and knowing in detail in which real conditions other algorithms have been performed. Having a Web-based tool to configure and execute benchmarks opens the door to new opportunities as the design of virtual tele-laboratories that permit the implementation of new algorithms using specific and detailed constraints. This is fundamental for designing benchmarks that allow the experiments to be made in a more scientific manner, taking into account that these experiments should be able to be reproduced again by other people under the same circumstances. In the context of underwater interventions with semi-autonomous robots, the situation gets even more interesting. Experiments performed by other researchers normally do not include the whole set of real conditions such as visibility or even water currents data that would allow the best scientific procedure. Underwater interventions and specially those performed on real sea scenarios are expensive, difficult to perform and reproduce. The contribution of this work is an online configuration tool for benchmarking using UWSim simulator which can provide an enormous help for this particular scenario, as will be shown in this paper.