Intern - France

6 months Internship – Setup Humanoid Dexterous Hand

Overview/Job Summary

We are looking for a motivated intern to contribute to the development of a dexterous humanoid hand. The project combines mechanical assembly, embedded electronics, multimodal sensing, physics simulation and robotic control.
The first phase will focus on understanding existing humanoid-hand platforms, including solutions such as Robot Nano Hand and Orca Hand, and assembling a functional robotic hand. The intern will then develop a simulation and sensor-control pipeline before comparing classical direct-drive or model-based control with AI-based hand control.

Internship Description

- As part of the Software Engineering Team in Grenoble, your goal will be first to go through state-of-the art and study existing humanoid-hand architectures, actuators, transmissions and sensors.
- Bring-up hand hardware and software including assembly of motors, encoders, control electronics, physical simulators.
- Calibrate joints and verify communication, limits and safety functions.
- Integrate multimodal sensors such as joint encoders, motor current, force/tactile sensors and cameras.
- Develop a physics-based simulation using tools such as MuJoCo, Isaac Sim, Gazebo or PyBullet.
- Build a software pipeline for sensor acquisition, synchronization, filtering, logging and control.
- Implement baseline controllers, including position, velocity, torque, impedance or synergy-based control.
- Collect data and develop an AI-based controller using imitation learning, reinforcement learning or residual learning.
- Deploy and test the AI model on the physical hand where possible.
- Compare classical and AI-based approaches using tracking accuracy, grasp success, latency, robustness, energy consumption and safety.

Qualifications

- Basic knowledge of robotics, kinematics and feedback control.
- Understanding of PID, position, velocity or torque control.
- Strong Python programming skills; basic C++ knowledge is desirable.
- Familiarity with Linux, Git and numerical tools such as NumPy and SciPy.
- Experience with or interest in ROS/ROS 2 and robotic simulation.
- Basic knowledge of machine learning and PyTorch or TensorFlow.
- Understanding of sensor calibration, filtering and data processing.
- Basic electronics and embedded-systems knowledge.
- Practical ability to assemble and test mechanical and electronic systems.

Duration: 6 months

American R&D Center (ARDC) is a central R&D organization of the TDK Group. Its mission is to contribute into the design of new products within TDK with the ambition to create new product lines. ARDC is working closely with InvenSense Inc., a TDK Group Company, who is the world’s leading provider of MEMS sensor platforms. InvenSense’s vision of Sensing Everything™ targets the consumer electronics and industrial markets with integrated Motion, Sound, Pressure, Gas solutions.

InvenSense’s motion tracking, audio, and services can be found in many of the world’s largest and most iconic brands including smartphones, tablets, wearables, drones, gaming devices, internet of things, automotive products, and remote controls for smart TVs.

Both ARDC & InvenSense are headquartered in San Jose, CA, and have offices in Boston, China, Taiwan, Korea, Japan, France, Slovakia, and Italy.