Actuator model, Identification and Differential Dynamic Programming for a TALOS humanoid robot

  • Posted on: 25 March 2021
  • By: ostasse
TitleActuator model, Identification and Differential Dynamic Programming for a TALOS humanoid robot
Publication TypeConference Paper
Year of Publication2020
AuthorsRamuzat, N, Forget, F, Bonnet, V, Gautier, M, Boria, S, Stasse, O
Conference NameEuropean Conference on Control

In this experimental paper, we would like to validate a non linear optimal control solver to realize torque controlon actuators embedded in a TALOS humanoid robot. Thetargeted application involves high payload, thus, it is necessaryto handle the mechanical limitations of the system. To thisextent, we propose a method to model, identify and control theTALOS humanoid actuators. The model includes the actuatordrive chain and the corresponding inertial parameters that areidentified at once using two experimental dataset. The identifiedmodel is then used by a Differential Dynamic Programming(DDP) optimal control solver to take into account the actuator limits. We demonstrated that the DDP can decrease the qualityof the tracking to avoid physical limits in angular position,velocity and current in extreme conditions such as carrying largeloads. Because of the solver high computational time, we validateour method on one actuator of the robot, the elbow joint, usingits main CPU. In the experiments, we charge up to 34kg on the arm of the robot at 5cm of the elbow joint, corresponding to 16N at the joint level. The proposed implementation isworking on this specific joint at 300μs and provide an effectivesolution to a real-world control problem. In the future, we willimplement it over dedicated and embedded electronics boardattached to each actuator.