Project description
Reinforcing smart autonomous vehicles for studying marine animals
One of our common European goals is the protection of the healthy and biodiversity of the marine ecosystem as the extremely important underwater environment is today in a danger. However, the protection and conservation of European marine waters need today completely new, ground-breaking approaches to achieve real improvements. The EU-funded AIforUTracking project will conduct cutting-edge research which focuses on the tracking of marine animals by autonomous vehicles using techniques of Reinforcement Learning (RL). New algorithms for more autonomy for machines will be designed thanks to novelty strategies and collaborations allowing applications like the Partially Observable Markov Decision Process (POMDP) or Multi-Agent Reinforcement Learning (MARL) to revolutionise possibilities of marine animal studying. Those methods will be tested and upgraded for better results.
Objective
The Artificial Intelligence methods for Underwater target Tracking (AIforUTracking) project will bring to the scientific community new tools for underwater target tracking by Autonomous Underwater Vehicles (AUVs) using Reinforcement Learning (RL) techniques. Moving towards the envisioned applications of marine animal tracking by autonomous vehicles, this proposal is clearly at the forefront of research, and directly addresses some of the main challenges and needs of the last Marine Strategy Framework Directive of the European parliament and of the Council, in particular establishing a framework for community action in the field of marine environmental policy. This research project will directly contribute to maintain and improve the health of the ocean by establishing innovative and unique research collaborations, and by introducing novel concepts and original research strategies that could provoke breakthroughs in the field of marine animal behavioural studies by:
a) Designing and developing optimisation algorithms that leverage new RL approaches, such as Partially Observable Markov Decision Process (POMDP) and Multi-Agent Reinforcement Learning (MARL). These Artificial Intelligence (AI) tools will increase the autonomy of the AUVs while improving the accuracy of the estimated target position.
b) Demonstrating the effectiveness and application of the path optimisation technique using POMPD and MARL methods by conducting real tests in the ocean, i.e. different targets will be tracked using a single AUV or multiple AUVs, as a proof-of-concept. These innovative technologies, together with Range-Only and Single-Beacon (ROSB) and Area-Only Target Tracking (AOTT) methods, are more competitive and offers greater autonomy than the traditional Long BaseLine (LBL) arrays-based methods.
Fields of science
Keywords
Programme(s)
Funding Scheme
MSCA-IF - Marie Skłodowska-Curie Individual Fellowships (IF)Coordinator
28006 Madrid
Spain