We are looking for motivated, enthusiastic and team-player colleagues in the area of robotics, biomechanics, neuroscience and AI.
We aim to offer an inclusive and friendly environment with a good work-life balance and livable salary. We also offer, in our opinion, the best place to work and live with the beautiful McGill Campus close to the Mont Royal in the heart of Montreal, the best city in Canada (and North America, some say).
To join us, please send an email to guillaume.durandau [at] mcgill.ca with in the name of the project and position in the message subject. You can also contact us if you have secured or to secure personal fellowship/funding for Postdoc/PhD/Master, just send an email with the funding and position in the message subject.
I do not reply to unsolicited email without project or funding in the email subject as well as not personalized email.
To maximize your chance of reply include in your email, why you want to join the lab and why your expertise and personality fit the lab.
You can check scholarship opportunities below:
- International Students Funding Opportunities
- McCall MacBain Scholarship for Master Students
- FRQNT Master Scholarship
- FRQNT PhD Scholarship
- FRQNT Postdoc Scholarship
- Banting Postdoc Scholarship
- Vanier PhD Scholarship
- Other Master funding
- Other PhD funding
- Other Posdoc funding
The available project and position are the following:
We have three undergrade summer intership opportunity:
nternship 1: Simulation of a neuromusculoskeletal model of the back with an exoskeleton
Back exoskeletons are one of the main research axes for reducing injury in the workforce but it is unknown what the effect of mechanical force on the spine and how to design useful exoskeletons for the back. This is due to the complexity of the spine neuromusculoskeletal system which requires multiple design iterations and tests on real subjects. One way to accelerate the design of back exoskeleton and create efficient and useful assistance for reducing injury is by simulating both the spine neuromusculoskeletal system and exoskeleton and looking at how they influence each other. This is why this project aims to develop a new back and exoskeleton model in MyoSuite (https://sites.google.com/view/myosuite) and validate it against experimental data previously collected. All results and developed models and code will be made open-source at the end of the project.
Internship 2: Reinforcement learning-based control of a neuromusculoskeletal model of the back with an exoskeleton
Back exoskeletons are one of the main research axes for reducing injury in the workforce but it is unknown what the effect of mechanical force on the spine and how to design useful exoskeletons for the back. This is due to the complexity of the spine neuromusculoskeletal system which requires multiple design iterations and tests on real subjects. One way to accelerate the design of back exoskeleton and create efficient and useful assistance for reducing injury is by simulating both the spine neuromusculoskeletal system and exoskeleton and looking at how they influence each other. This is why this project aims to develop a policy trained using reinforcement learning for the control of neuromusculoskeletal models wearing a passive exoskeleton. The policy will be able to reproduce unknown (i.e. not used for the training or the reward shaping) experimental data previously recorded by just following the kinematics cue.
Internship 3: Studying step response in balance using reinforcement learning
Fall is one of the main contributors to loss of mobility in the older population. Injuries caused by falls can reduce the quality of life and even cause death. Defining a new robotic device and controller for increased balance recovery and preventing falls will have a huge impact int he quality of life in the elderly population. Unfortunately, testing devices on the targeted groups is challenging. For this neuromusculoskeletal predictive simulations have been developed where experiments can be conducted in simulation. We previously developed a neuromusculoskeletal model of the lower limb driven by a policy trained using reinforcement learning that can sustain low perturbation. We would like to look at the possibility of sustaining higher perturbation and generating a stepping response. Stepping response is a balanced recovery strategy where we step forward to regain stability. Creating simulations that use natural and biological balance strategies is a step forward towards the creation of better predictive stimulation that will help the elderly population regain autonomy.
More info on how to apply: https://www.mitacs.ca/en/programs/globalink/globalink-research-internship?utm_sourc[…]medium=SocialMedia&utm_campaign=GRIStudentCall2024-SocialMedia
To look at the internship's page go to: https://globalink.mitacs.ca/#/student/application/projects and in Faculty last name put: Durandau
These internships are funded by Mitacs and participants in the Mitacs internship program will be eligible for a Globalink Graduate Fellowship for doing a Master's or PhD in Canada: https://www.mitacs.ca/en/programs/globalink/globalink-graduate-fellowship
/!\ Deadline for application End of September 2023 /!\