References of "Ludivig, Philippe 50035262"
     in
Bookmark and Share    
Full Text
Peer Reviewed
See detailABSOLUTE LOCALIZATION FOR SURFACE ROBOTICS IN GPS-DENIED ENVIRONMENTS USING A NEURAL NETWORK.
Wu, Ben; Ludivig, Philippe UL; Potter, Ross et al

Poster (2020, October 23)

Accurate localization in surface robotics is essential for navigation, path planning, and science objectives. On Earth, absolute localization can be readily achieved via satellite navigation (e.g., GPS ... [more ▼]

Accurate localization in surface robotics is essential for navigation, path planning, and science objectives. On Earth, absolute localization can be readily achieved via satellite navigation (e.g., GPS). For other planetary bodies such as the Moon or Mars, however, such systems are unavailable. Current methods for absolute localization of planetary rovers rely on time- and labor-intensive human visual matching of surface perspective features with satellite images. Relative localization also accumulates errors over time, with different methods estimating dissimilar locations (e.g., [1]). Thus, an absolute localization method that can quickly, automatically, and accurately reduce the position search space is of great benefit to future planetary exploration missions. This project [2] presents a new approach to localizing planetary rovers: training an artificial neural network to match surfaceperspective imagery to corresponding satellite maps. [less ▲]

Detailed reference viewed: 41 (1 UL)
Full Text
Peer Reviewed
See detailLarge scale realistic virtual environments for lunar robotics testing using real-time computer games engines
Ludivig, Philippe UL; Voos, Holger UL; Lamamy, Julien

Poster (2020, October 23)

Virtual simulation environments are a great tool for testing different rover systems before they are sent to the moon. They allow for perfect repeatability and give abetter idea how specific parameters ... [more ▼]

Virtual simulation environments are a great tool for testing different rover systems before they are sent to the moon. They allow for perfect repeatability and give abetter idea how specific parameters can impact the overall system. Most current simulators however suffer from low visual fidelity which is problematic for testing the vision sensors needed for autonomous navigation.Additionally, due to technical and practical limitations, most simulations are limited in environment scale. This is problematic for long range navigation testing needed for missions such as the mars sample fetch rover[1]. In order to solve both issues, we turn towards a popular computer game engine,Unreal Engine 4. It allows for larger environments than what is currently possible in robotics simulators such as Gazebo[2]or V-rep[3]. It also addresses the visual fidelity with a range of tools including real-time ray-tracing.The tradeoff is a low physics fidelity, which can be an issue when testing wheel-soil interaction. For our use-case we focus primarily on perception systems needed for rover navigation, such as mono and stereo camera systems, where visual fidelity is more important. [less ▲]

Detailed reference viewed: 60 (7 UL)
Full Text
Peer Reviewed
See detailTESTING ENVIRONMENTS FOR LUNAR SURFACE PERCEPTION SYSTEMS; COMBINING INDOOR FACILITIES, VIRTUAL ENVIRONMENTS AND ANALOGUE FIELD TESTS.
Ludivig, Philippe UL; Olivares Mendez, Miguel Angel UL; Calzada Diaz, Abigail et al

Scientific Conference (2020, October 21)

This paper describes the different approaches which can be used to test vision systems for operations on robotic lunar surface missions. We investigate validating systems in virtual environments, lab ... [more ▼]

This paper describes the different approaches which can be used to test vision systems for operations on robotic lunar surface missions. We investigate validating systems in virtual environments, lab environments and analogue outdoor environments and demonstrate that a combination of all three approach-es is needed to sufficiently test systems for the lunar surface. [less ▲]

Detailed reference viewed: 45 (6 UL)
Full Text
Peer Reviewed
See detailBUILDING A PIECE OF THE MOON: CONSTRUCTION OF TWO INDOOR LUNAR ANALOGUE ENVIRONMENTS
Ludivig, Philippe UL; Calzada-Diaz, Abigail; Olivares Mendez, Miguel Angel UL et al

in Proceedings of the 71st International Astronautical Congress 2020 (2020, October 12)

Developing and testing autonomous systems to ensure that they work reliably on the moon is a difficult task, as testing on location is not an option. Instead, engineers rely on simulations, testing ... [more ▼]

Developing and testing autonomous systems to ensure that they work reliably on the moon is a difficult task, as testing on location is not an option. Instead, engineers rely on simulations, testing facilities and outdoor lunar analogues. Due to the lack of lunar analogue testing facilities in Europe, ispace Europe and the University of Luxembourg have teamed up to build two of these facilities with the goal of designing new vision-based navigation systems. These systems will enable autonomous long-range traverses for lunar rovers. These two facilities have a surface area of 64 and 77 square meters, respectively. Regarding the type of testing needed for vision-based systems, the optical fidelity of the environment has been considered as the most important factor. Thus, different types of Basalt have been used for the two facilities to create a larger number of possible landscapes, such as craters, hills, rocky areas and smooth planar surfaces. Regolith simulant was also considered but, due to the health restrictions and the cost factor, basalt was selected instead. As a result, this has allowed for larger testing areas. The illumination setup has been designed to simulate the highland regions of the Moon, with a single light source positioned low above the horizon, casting long shadows over the entire area. To mitigate problems with feature detection algorithms picking up features at the edge of the facility, the walls have been painted black. This also produces high contrast shadows, which is exactly what makes vision-based navigation challenging in the polar regions. The outcome of this research is a set of lessons learned which will enable other researchers to replicate similar facilities and to reproduce the same fidelity in indoor testing for future vision-based navigation systems. [less ▲]

Detailed reference viewed: 83 (17 UL)