No abstract

Scaling end-to-end reinforcement learning to control real robots from vision presents a series of challenges, in particular in terms of sample efficiency. Against end-to-end learning, state representation learning can help learn a compact, efficient and relevant representation of states that speeds up policy learning, reducing the number of samples needed, and that is easier to interpret. We evaluate several state representation learning methods on goal based robotics tasks and propose a new unsupervised model that stacks representations and combines strengths of several of these approaches. This method encodes all the relevant features, performs on par or better than end-to-end learning with better sample efficiency, and is robust to hyper-parameters change.

We propose a fault-tolerant estimation technique for the six-DoF pose of a tendon-driven continuum mechanisms using machine learning. In contrast to previous estimation techniques, no deformation model is required, and the pose prediction is rather performed with polynomial regression. As only a few datapoints are required for the regression, several estimators are trained with structured occlusions of the available sensor information, and clustered into ensembles based on the available sensors. By computing the variance of one ensemble, the uncertainty in the prediction is monitored and, if the variance is above a threshold, sensor loss is detected and handled. Experiments on the humanoid neck of the DLR robot DAVID, demonstrate that the accuracy of the predicted pose is significantly improved, and a reliable prediction can still be performed using only 3 out of 8 sensors.

No abstract

In contrast to underactuated robotic hands the DLR AWIWI II hand of the David robot is fully controllable because each finger with 4 joints is actuated by 6 or 8 tendons respectively. For such fingers all joint angles (generalized positions) or joint torques (generalized forces) can be controlled independently. Usually, the specifications in joint space are converted to desired tendon forces or motor torques, which are regulated by an inner loop impedance controller. However, this conversion typically exhibits couplings between the components of the joint angle vector or the joint torque vector respectively, which arise when using the well known equations. Therefore the usual force control and position control schemes are reviewed and a generic computation of the desired tendon forces is presented. This is also done for the control of the Cartesian position and force at the finger endpoint. Thus the main contribution of the paper is the inhibition of couplings in joint space or at the Cartesian endpoint. This is demonstrated in simulations of the index finger of the DLR David hand.