Dynamics, System Identification & Estimation Engineer - PLA (human)

Metzingen / Riederich. Posted 5 months ago

The role

This person will own dynamic model accuracy and system identification on 4NE1 hardware. They will author simulation models in MuJoCo and Isaac Sim and implement real-time state estimation.

What you will do

Define model fidelity metrics and own the gap between simulation and real-hardware behaviour
Perform system identification on 4NE1 hardware including motor constants and joint friction
Author and maintain MuJoCo and Isaac Sim models
Implement and tune floating-base EKF/UKF for real-time state estimation
Develop sim-to-real pipelines using parameter estimation loops and hardware-data-driven calibration
Debug model-accuracy-driven failures such as control instability or estimation drift
Supply updated Pinocchio model parameters to Core Robot Software engineers

Required qualifications

MSc or PhD in Robotics, Mechanical Engineering, Electrical Engineering, or a related field
4+ years of experience developing state estimation or system identification solutions for real-time robotic systems on real hardware
Experience with system identification on physical robotic systems
State estimation implementation using EKF or UKF for floating-base pose, velocity, and contact state on a legged or mobile robot platform
Depth in rigid body dynamics including contact modelling and actuator behaviour
Experience supporting control systems (MPC, WBC) or learned policies (RL) through hardware deployment
C++ for production RT systems
Python for analysis, tooling, and calibration pipelines

Nice to have

Humanoid or legged robot hands-on experience
Differentiable simulators for gradient-based system identification (Brax, DiffTaichi, or comparable)
Sim-to-real transfer methodology: domain randomisation, adaptive calibration, residual physics modelling
Pinocchio for rigid-body model computation and parameter sensitivity analysis
MuJoCo model authoring: MJCF contact parameters, actuator models, tendon dynamics
Factor graph-based estimation (GTSAM, iSAM2) for tightly-coupled IMU + kinematics fusion
Publications or open-source contributions in legged robot dynamics, system identification, or sim-to-real transfer

About the company

neura-robotics-gmbh