Robotics Moonshot : digital twin of a laser cutting process and implementation with a self-learning robot
One of the main challenges in the deployment of robotics in industry is to offer smart robots, capable of understanding the context in which they operate and easily programmable without advanced skills in robotics and computer science. In order to enable a non-expert operator to define tasks subsequently carried out by a robot, the CEA is developing various tools: intuitive programming interface, learning by demonstration, skill-based programming, interface with interactive simulation, etc.
Winner of the "moonshot" call for projects from the CEA's Digital Missions, the "Self-learning robot" project proposes to bring very significant breakthroughs for the robotics of the future in connection with simulation. A demonstrator integrating these technological bricks is expected on several use cases in different CEA centers.
This post-doc offer concerns the implementation of the CEA/DES (Energy Department) demonstrator on the use case of laser cutting under constraints for A&D at the Simulation and Dismantling Techniques Laboratory (LSTD) at the CEA Marcoule.
Elaboration of a common robot/human action space
This post-doc aims at establishing by artificial intelligence methods (e.g. signal processing on graphs), the mapping of an industrial task performed by a human operator, and acquired by visual sensors, in order to be interpretable and exploitable by a robot. It is part of a project aiming at designing a demonstrator in which a robot will learn to reproduce by observation a task performed by a human. The platform has been deployed at CEA Tech and is currently operated by an engineer.
The objective of this post-doc is mainly to study and develop a set of methods to build a mapping between the actions performed by a human operator and perceived through visual sensors and the actions performed by the robot. These methods and the work of the related theses will then be implemented in the demonstrator in order to test them experimentally.
Due to the central position of the subject of this post-doc, under the triple supervision of the PACCE and IPI teams of LS2N and CEA, you will have to collaborate closely with the two PhD students already involved in the project. You will have to conceptualize and formalize the methods and representations on the one hand by synthesizing the existing literature on the subject and on the other hand by establishing a common framework encompassing the two thesis works.
Apprenticeship Learning Platform deployment for industrial applications
This project aims at developing a demonstrator that integrates state-of-the-art technologies and improve it on a use-case representative of the industrial world.
The demonstrator will consist in a robotic / cobotic arm coupled to an acquisition sensor (RGBD type). This device will be positioned in a workspace made of a rack / shelf containing objects / pieces of various shapes and qualities (materials, densities, colors ...) in front of which will be placed a typical conveyor prototype of industrial installations. The type of tasks expected to be carried out by the demonstrator will be "pick and place" type tasks where an object will have to be identified in shelf and then placed on the conveyor.
This type of demonstrator will be closer to the real industrial conditions of use than the "toy" examples used in the academic field.
This demonstrator will focus first on the short-term effectiveness based on state of the art technologies for both hardware and software, for a use case representative of the industrial world.
At first, it will thus be less focused on the evolution of the algorithms used than on the adaptation of the parameters, the injection of knowledge a priori dependent on the context making it possible to reduce the high-dimensional input space, etc.
Design and implementation of a bio-inspired sense, application to offshore teleoperation and to operator assistance
In recent years, the Bio-inspired Robotics Group of Robotics team IRCCyN has developed an artificial bio-inspired electric fish sense. To emulate the electrical sense, resistive probes were used for piloting the IRCCyN submarine autonomous robot.
For its part, within the Interactive Robotics Laboratory (LRI), the CEA LIST has been pursuing for several years a research activity in the field of force feedback telerobotics. The operator manipulates a slave robot located in hostile environments via a master arm located in a safe area and a computer system.
The candidate’s work will take place in a CEA- IRCCyN project running in parallel over a first project whose purpose is to demonstrate the concept of electro- haptic loop on a Cartesian arm carrying an electric probe with a fixed and known geometry. The postdoc will be in charge of implementing the loop on a "marinized" manipulator arm with a complex geometry. To do this, with the assistance of CEA and IRCCyN , he/she will support the preparation of this arm and adaptation of electrical sensor (emitter electrodes , receiver , electronic) architecture considered , as well as the adaptation of the monitoring / control of the haptic interface at the base of the electro-haptic loop. In addition to the technological challenges of this adaptation, the candidate must also consider different strategies to exploit the electric field on a multi-body system of variable geometry.
Experimental validation and proof of concept of this new offshore teleoperation system will be carried out on scenarios, to be defined, representative of the final application.
Design and implementation of force feedback by electrical sense for remote operation with submarines or aerial robots
Since few years, the Bio-inspired Robotics group of the IRCCyN Robotics team is developing a bio-inspired perception mode found on some freshwater tropical fish: the electrical sense. This active sense is based on the distortions measures, due to environment, of an electric field produced by the fish. Based on this principle Irccyn developed in the context of a European project called Angels, the first autonomous underwater robot capable of moving by means of the electrical sense . In the future, CEA TECH and Irccyn want to extend this first result in multiple directions, including the remote operation of submarines manipulators and aerial robots domains. The force feedback should be emulated by the use of the electrical sense. Integrated in the Bio-inspired robotics team of IRCCYN , post -doctoral fellow will contribute to the development of the electric sense and its use for underwater and aerial teleoperation . He will participate in the design and development of new sensors inspired by electric fish and their use for underwater telerobotics. The results of its work will underpin the industrial demonstrator system (teleoperation offshore) to be developed under the project CEA TECH / IRCCYN Bio-inspired robotics.
Realistic motion generation for anthropomorphic systems
This work is devoted to Digital Human motion generation, for manufacturing (specifically, for design, maintenance, operator training, workstation design and ergonomics, ...), health (postures for surgery, rehabilitation, ...), or entertainment industry (animation for games, movies,...).
Based on complementary skills and developments of Gepetto LAAS team, and CEA LIST team, in terms of path planning (HPP), dynamic motion control of anthropomorphic systems, the objective of this post doc is to combine both approaches, a global one, dealing mainly with geometric and quasi-static constraints and characteristics, and a local one, dealing with dynamics and taking into account human movement characteristics (motor primitives, minimizing cost criteria, etc ...).
Simultaneous Localisation and Mapping with an RGB-D camera based on a direct and sparse method
Recent advances in the methods of locating a device (smartphone, robot) in relation
to its environment make it possible to consider the deployment of augmented reality solutions and autonomous robots. The interest of RGB-D cameras in such a context is notable since it allows to directly acquire the depth map of the perceived scene.
The objective of this post docorate consists in developping a new SLAM (Simultaneous Localisation and Mapping) method relying on a depth sensor.
To reach a solution both robust, accurate and with small CPU/memory comsumption, the depth image will be exploited though a direct and sparse approach. The resulting solution will be then combined with the solution of "RGB SLAM Constrained to a CAD model" developped in our laboratory, resulting finaly in an "RGB-D SLAM Constrained to a CAD model"
modelling and Control of voltage and frequancy in GALS architecture submitted to Process-Voltage-Temperature variability
The evolution of sub-micron technologies has induced tremendous challenges the designer has to face, namely, the Process-Voltage-Temperature varibility and the decrase of power consumption for mobile applications.
The work to be done here concerns the DVFS (Dynamic Voltage and Frequency Scaling) policies for GALS (Globally Asynchronous, Locally Synchronous) architecture.
A fine grain modelling of the voltage and frequency “actuators” must be first done in order to simulate in a realistic ways the physical phenomena. Especially, the various parameters that may influence the system will be considered (process variation, supply voltage variation and noise, temperature variation, etc.)
Then, Non-Linear (NL) control laws that take into account the saturation of the actuators will be developed. These laws will be validated on the physical simulator and their performances in regulation (i.e. the response of the closed-loop system to disturbances such as PVT variations) will be evaluated. Note that these laws will be designed at the light of implementation constraints (mainly cost) in terms of complexity, area, etc.
Actually, the system considered here is intrinsically a Multi-Inputs-Multi-Outputs (MIMO) one. Therefore, its control can be design with NL techniques devoted to MIMO systems in order to ensure the requirements and reject the disturbances.
The control of several Voltage and Frequency Islands (VFI) is usually done via a “central brain” that chooses the voltage and frequency references thanks to a computational workload deadline. For more advanced architectures, the capabilities of each processing element, especially its maximum frequency, can be taken into account. A disruptive approach should be to consider a more distributed control that for instance takes into account the particular state (e.g. temperature) of each VFI neighbours. Control techniques that have been designed for distributed Network Controlled Systems could be adapted to MPSoCs.
Design of a control system of a plane based on distributed electric propulsion
The objective of this post doctorate is to design a control system to manage the electrical power on an electric plane proposed by many electrical turbines. The aim of the work is to demonstrate the possibility to increase the propulsion efficiency by using many cooperative electrical turbines placed judiciously on plane compare to a plane having only two or four turbine. Furthermore, one idea is to completely drive the plane by adjusting in real time the power of each electrical turbine taking advantage of their high reactivity compare to classical thermal turbines. The background required for that post doctorate is a good knowledge in control system and power electronic.