Objectives: The main objective of this thesis is to design, build, and initiate the validation of a robotic knee simulator capable of imitating a walking motion in an automated manner. This device is intended to serve as a calibration tool to validate angular measurement methods using inertial sensors.
Methodology: The project began with the complete mechanical and electrical design of the simulator, including material selection, the integration of five electric motors, and the implementation of a control architecture based on Matlab/Simulink and a Raspberry Pi microcontroller. Communication between the controller and the motors was achieved through the CAN protocol. A Simulink model was developed to manage the generation of the hip and knee angular displacement signals during walking, the simulator’s homing sequence, and the execution of the walking profile. Finally, experimental verification was conducted in two stages: an internal verification based on the motors, CAN feedback, followed by an external verification using optoelectronic motion capture (OptiTrack).
Results: The results showed that the simulator can imitate the programmed walking angular trajectories with accurate position tracking. Internal verification confirmed the consistency between the transmitted commands and the motor-measured positions, with a mean absolute error below 1° and a maximum deviation of 2°. The external verification, performed using the OptiTrack system, allowed comparison between the simulator’s angular displacements and reference biomechanical angular data, confirming good overall agreement between the walking profiles. Some artifacts were observed, mainly related to CAN transmission, but they did not significantly affect the general shape of the angular trajectories (variation below 1% of the total amplitude).
Conclusion: This thesis demonstrated the mechanical feasibility of a robotic knee simulator for the controlled reproduction of walking movements. The strength and stability of the structure were evaluated through torque and stress analyses performed in SolidWorks, confirming the structure’s ability to withstand motor-generated loads without significant deformation. The device provides a reproducible protocol and represents a relevant platform for calibrating and validating inertial sensors or any other measurement system. Several opportunities for improvement were identified, particularly regarding the reduction of transmission errors and software optimization, in order to enhance the simulator’s precision and reliability.
Vincent, X. (Author),
Hagemeister (Supervisor) &
Aissaoui (Co-supervisor),
10 Dec 2025Student thesis: Master's thesis › Master in Engineering: Engineering