- Satya Chavan MARTHI successfully defended his Masters thesis title: Safe Reinforcement Learning for Discrete-Time Control Affine Non-linear Systems under Input Saturation" under my supervision. He will now join our group for PhD thesis on the subject: Design of a safe control system through reinforcement learning - Application to autonomous mobile systems.
under the supervision of Didier Theillol and Mayank Jha.
Department: CRAN-CID,
Funding: ANR SOS Project (Self-Organizing, Smart, and Safe Heterogeneous Robots Fleet by Collective Emergence),
Consortium: CRAN, CRISTAL (Lille), and Lynxdrone (SME)
Link: https://projet-sos.gitlabpages.univ-lille.fr/site-web/
- Mayank JHA co-organized, along with Vasso Reppa (Department of Maritime and Transport Technology, Delft University of Technology, TU Delft, Netherlands) and Didier Theilliol, a roundtable discussion on the topic:“Gnosis for Maintenance: From Diagnosis to Prognosis and Health-Aware Control”, featuring expert panelists Alfredo Núñez Vicencio (Associate Professor at Delft University of Technology, Netherlands), Jin Jiang (Professor at the University of Western Ontario, Canada), and Vicenç Puig (Professor at the Technical University of Catalunya (UPC), Spain).
This discussion took place at SAFEPROCESS 2024 in Ferrara, Italy, in June 2024.
The experts shared their insights and experiences, leading to high-quality discussions and scientific exchanges on various topics, including prognostics, diagnostics, control reconfiguration, health-aware control, safe control learning, and their mutual interactions! See the details here.
- Mayank JHA, along with Vasso Reppa (TU Delft) and John Jairo Molina (GIPSA Lab, Grenoble), has proposed the creation of a ‘Working Group’ under the TC 6.4 (SAFEPROCESS) of IFAC titled: “Health-Aware Control and Safe Control Learning for Safety-Critical Systems”.
The working group aims to bring together researchers from the domains of Prognostics, Dependability, Control Theory, and Reinforcement Learning.
It will encourage activities such as the organization of invited/special sessions at IFAC conferences, and the publication of special issues in prestigious journals such as Annual Reviews in Control, IEEE Control Systems Technology, etc. -
Mayank JHA was invited to the NASA Ames Research Center, located in Silicon Valley, Mountain View, USA, in June 2023.
On this occasion, Mayank JHA delivered a one-hour lecture on the topic: “Safe Reinforcement Learning and System Identification with Deep Learning for Prognostics” to NASA researchers.
His presentation addressed issues related to safe control learning as well as the design of health-aware control for critical systems.
Had the opportunity to visit NASA’s Prognostics Center of Excellence and engage in discussions on various collaborative areas within the field of prognostics and control.
Following this visit, at the end of June 2023, Chetan Kulkarni, KBR Technical Fellow at KBR, Inc at NASA Ames Research Center, visited CRAN in Nancy, France.
During his visit, Mr. Kulkarni gave a talk on the topic: “Model-Based Approaches for Fault Detection, Prognostics, and Decision-Making for Complex Systems.” In his presentation, he highlighted the outstanding work of his team at NASA Ames, particularly in the fields of diagnostics, prognostics, and physics-informed neural networks (PINNs).
