DIOMEDES
 | Subprogram: RETOS DE LA SOCIEDAD ANDALUZA Reference: UMA18-FEDERJA-090 Title: DESARROLLO DE TÉCNICAS DE CONTROL INTELIGENTE CON APRENDIZAJE PARA NAVEGACIÓN DE VEHÍCULOS AUTÓNOMOS EN ENTORNOS NO ESTRUCTURADOS (DIOMEDES) Principal Investigator 1: FERNÁNDEZ LOZANO, JUAN JESÚS Principal Investigator 2: MORALES RODRÍGUEZ, JESÚS |
Summary
Mobility of people and goods is one of the challenges with the greatest potential impact in the coming years, due to its economic, environmental and safety implications. In all three aspects, vehicle autonomy has taken a central role in the expected evolution of transport. Thanks to various advances in the development of intelligent control systems, autonomous vehicles are beginning to emerge and test on the road. But there are applications of high economic and societal impact where the development of these intelligent control systems is more demanding, and have received comparatively less attention. This is the case for applications in less structured environments, such as the natural environments found in agricultural or search & rescue applications, or the complex environments of logistics or industrial centres.
The project's mission is to develop intelligent control systems for the navigation of autonomous vehicles in unstructured environments, in the presence of other vehicles and humans. These navigation systems must allow the vehicle to move safely from an initial point to an end point within its range, choosing the best path to do so (global navigator), avoiding natural obstacles, people or other vehicles in the area (local navigator). A methodology will be developed based on the application of intelligent control techniques supported by their adjustment through learning in simulated environments, and their transfer to a real application. As a reference application, search and rescue in natural environments is proposed, for which an experimental field is available, as well as various mobile robots and sensors of different types, which will allow the experimental validation of the results of this project under realistic conditions.
Publications
Martínez, J.L.; Morán, M.; Morales, J.; Robles, A.; Sánchez, M. Supervised Learning of Natural-Terrain Traversability with Synthetic 3D Laser Scans . Appl. Sci. 2020 , 10 , 1140. https://doi.org/10.3390/app10031140
Martínez, J.L.; Morales, J.; Sánchez, M.; Morán, M.; Reina, A.J.; Fernández-Lozano, J.J. Reactive Navigation on Natural Environments by Continuous Classification of Ground Traversability . Sensors 2020 , 20 , 6423. https://doi.org/10.3390/s20226423
Morales J, Vázquez-Martín R, Mandow A, Morilla-Cabello D, García-Cerezo A. The UMA-SAR Dataset: Multimodal data collection from a ground vehicle during outdoor disaster response training exercises . The International Journal of Robotics Research . 2021 ;40(6-7):835-847. https://doi.org/10.1177/02783649211004959
Morales, J.; Martínez, J.L.; García-Cerezo, A.J. A Redundant Configuration of Four Low-Cost GNSS-RTK Receivers for Reliable Estimation of Vehicular Position and Posture . Sensors 2021 , 21 , 5853. https://doi.org/10.3390/s21175853
Sánchez, M.; Morales, J.; Martínez, J.L.; Fernández-Lozano, J.J.; García-Cerezo, A. Automatically Annotated Dataset of a Ground Mobile Robot in Natural Environments via Gazebo Simulations . Sensors 2022 , 22 , 5599. https://doi.org/10.3390/s22155599
Morales, J.; Castelo, I.; Serra, R.; Lima, P.U.; Basiri, M. Vision-Based Autonomous Following of a Moving Platform and Landing for an Unmanned Aerial Vehicle . Sensors 2023 , 23 , 829. https://doi.org/10.3390/s23020829
- Supplementary material (videos): Simulated following and landing test , real-world following test , real-worl landing test .
