Internship Engineering : Learning physical quantities on real robots using a Motion Capture system

Inria
February 15, 2023
Contact:N/A
Offerd Salary:Negotiation
Location:N/A
Working address:N/A
Contract Type:Other
Working Time:Negotigation
Working type:N/A
Ref info:N/A

2023-05696 - Internship Engineering : Learning physical quantities on real robots using a Motion Capture system

Contract type : Internship

Level of qualifications required : Master's or equivalent

Fonction : Internship Engineering

Context

Motion Capture systems allow for precise (absolute) measurements of objects position, orientation and velocity in the 3D world. Beside being able to track human motion (as it is mainly used in the cinema/video game industry), it is very convenient for robotics application, such as gathering “ground truths” during real-world experiments. Without the external motion capture system's measurements, robots can only rely on their onboard sensors to determine their position/orientation/velocity.

Although Motion Capture Systems are commonly used in the development of new methods, those methods cannot rely on the Motion Capture measurements once released, since this system is impractical to deploy in most environments.

Assignment

The goal of this internship is to create simple ML models to infer specific data learned from proprioceptive sensors using Motion Capture measures as ground truth during training.

Two different topics, based on two different robots, are available for this internship. It will be up to the intern to decide to focus only on one or to pursue both.

  • Contact detection between feet and the ground on a quadruped robot (Solo1212) using joint measurements (position, velocity ant torque) and IMU measurements.
  • Position error of the end effector of a Tiago3 robot, due to slack and flexibility in the joints, using only joint values.
  • Potential continuity of these works could be to:

  • For the contact detection on Solo-12 :
  • Reduce this model to the minimum to implement it on a lightweight microcontroller, embed it on the robot, and run it in real-time.
  • And/Or implement an Invariant Extended Kalman Filter using this contact detector.
  • For Tiago calibration :
  • Create a ros-package that uses this model to anticipate the error and plan a trajectory that account for it.
  • Connect this work with visual servoing
  • Main activities

    Despite the differences in the studied quantities, the workflow will be quite similar for both topics (from a global point of view).

  • Get familiar with the softwares developed by the team and learn how to use robots and Motion Capture system.
  • Generate meaningful datasets :
  • Generate motions to execute on the real robot (that are safe but expressive enough)
  • collect relevant data
  • filter and post process.
  • Choose and train an appropriate ML solutions
  • Characterize their performances and think ciritcally.
  • Skills

    Technical :

  • C++/Python
  • Basics in robotics: modelling, kinematics, dynamics, low-level control
  • Machine learning: scikit-learn, PyTorch
  • System: Linux, ROS, conda
  • (if you lack some of the above skills, this internship can be an opportunity for you to learn them)

    Languages :

  • French
  • English
  • Benefits package
  • Subsidized meals
  • Partial reimbursement of public transport costs
  • Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
  • Professional equipment available (videoconferencing, loan of computer equipment, etc.)
  • Social, cultural and sports events and activities
  • Access to vocational training
  • General Information
  • Theme/Domain : Optimization, machine learning and statistical methods Software Experimental platforms (BAP E)

  • Town/city : Paris

  • Inria Center : Centre Inria de Paris
  • Starting date : 2023-03-20
  • Duration of contract : 6 months
  • Deadline to apply : 2023-02-15
  • Contacts
  • Inria Team : WILLOW
  • Recruiter : Arlaud Etienne / etienne.arlaud@inria.fr
  • The keys to success

    We are looking for strongly motivated candidates with an interest in machine learning and robotics. The project requires a strong background in robotics and excellent programming skills.

    About Inria

    Inria is the French national research institute dedicated to digital science and technology. It employs 2,600 people. Its 200 agile project teams, generally run jointly with academic partners, include more than 3,500 scientists and engineers working to meet the challenges of digital technology, often at the interface with other disciplines. The Institute also employs numerous talents in over forty different professions. 900 research support staff contribute to the preparation and development of scientific and entrepreneurial projects that have a worldwide impact.

    Instruction to apply

    Defence Security : This position is likely to be situated in a restricted area (ZRR), as defined in Decree No. 2011-1425 relating to the protection of national scientific and technical potential (PPST).Authorisation to enter an area is granted by the director of the unit, following a favourable Ministerial decision, as defined in the decree of 3 July 2012 relating to the PPST. An unfavourable Ministerial decision in respect of a position situated in a ZRR would result in the cancellation of the appointment.

    Recruitment Policy : As part of its diversity policy, all Inria positions are accessible to people with disabilities.

    Warning : you must enter your e-mail address in order to save your application to Inria. Applications must be submitted online on the Inria website. Processing of applications sent from other channels is not guaranteed.

    From this employer

    Recent blogs

    Recent news