Internship: Decentralized learning of locomotion in modular soft robots
The Soft Robotic Matter group at AMOLF is looking for interns at the Master level from a range of backgrounds (including physics, mathematics and engineering) that will work on the development and understanding of decentralized learning strategies in modular soft robots. One of the main challenges in modular robotics is developing systems that can adapt to their environment and achieve autonomous behavior. Current approaches often accomplish this by increasing the complexity of the centralized controllers. In contrast, complex group behavior in nature often arises from relatively simple local interactions (e.g. bird flocking behavior, reaction-diffusion). Our project aims to simplify decentralized controllers to find the minimum requirements for a universal and scalable learning strategy in modular systems.
Previous experiments in the Soft Robotic Matter group with local stochastic optimization show great promise for emergent control in a collective one-dimensional locomotion task. We seek to build upon these results by increasing the complexity with experiments and simulations of two-dimensional grid configurations and three-dimensional body configurations of modular soft robots. Each module of the robot in the structure is a simple, physically connected unit cell that can independently actuate and interact with the environment through a microprocessor, servo, and light sensor. Using this setup, we aim develop robust and universal learning algorithms that can deal with various configuration and environments, in order to have the robot autonomously move towards a light source.
Internships could vary from computational work on the learning and control strategy in simulation to an experimental internship working with the currently developed modular robots or the construction of new modular robots, depending on the student’s background and interest.
About the group
The Soft Robotic Matter group focuses on the design, fabrication and fundamental understanding of materials that are capable of autonomously adapting to – and even harnessing – variations in their environment. We aim to uncover principles that help us understand how non-linearity and feedback can result in the emergence of complex – but useful – behaviour in soft actuated systems. To this end, the Soft Robotic Matter group explores active and sensing elements to implement feedback capabilities and computation in soft architected materials, and uses a combination of computational, experimental and analytical tools. This line of research uniquely combines concepts from soft robotics and architected materials, providing new and exciting opportunities in the design of compliant structures and devices with highly non-linear behaviour. See also: www.overvelde.com.
Qualifications
The internship is open to candidates from a range of backgrounds, including Engineering, Physics, Computer Science, Mathematics or a related field. The Soft Robotic Matter group is looking for a highly motivated candidate with a go-getter mentality. Excellent verbal and written communication skills (in English) are essential. The internship must be a mandatory part of your curriculum. You have a nationality of an EU-member state and/or you are a student at a Netherlands University. Please note: as from the 1st of January 2021 the UK is no longer an EU-member state. You must be available for preferably more than 6 months.
Terms of employment
At the start of the traineeship your trainee plan will be set out, in consultation with your AMOLF supervisor.
Contact info
Dr. ir. Bas Overvelde
Group leader Soft Robotic Matter Group
E-mail: overvelde@amolf.nl
Phone: +31 (0)20-754 7100
You can respond to this vacancy online via the button below.
Please annex your:
– Resume;
– List of followed courses plus grades.
Applications will be evaluated on a rolling basis and as soon as an excellent match is made, the position will be filled.
Online screening may be part of the selection.
AMOLF is highly committed to an inclusive and diverse work environment. Hence, we greatly encourage candidates from any personal background and perspective to apply.
Commercial activities in response to this ad are not appreciated.
