Technical session talks from ICRA 2012
TechTalks from event: Technical session talks from ICRA 2012
Conference registration code to access these videos can be accessed by visiting this link: PaperPlaza. Step-by-step to access these videos are here: step-by-step process .
Why some of the videos are missing? If you had provided your consent form for your video to be published and still it is missing, please contact email@example.com
Robust and Adaptive Control of Robotic Systems
A Nonlinear PI and Backstepping Based Controller for Tractor-Steerable Trailer Influenced by SlipAutonomous guidance of agricultural vehicles is vital as mechanized farming production becomes more prevalent. It is crucial that tractor-trailers are guided with accuracy in both lateral and longitudinal directions, whilst being affected by large disturbance forces, or slips, owing to uncertain and undulating terrain. Successful research has been concentrated on trajectory control which can provide longitudinal and lateral accuracy if the vehicle moves without sliding, and the trailer is passive. In this paper, the problem of robust trajectory tracking along straight and circular paths of a tractor-steerable trailer is addressed. By utilizing a robust combination of backstepping and nonlinear PI control, a robust, nonlinear controller is proposed. For vehicles subjected to sliding, the proposed controller makes the lateral deviations and the orientation errors of the tractor and trailer converge to a neighborhood near the origin. Simulation results are presented to illustrate that the suggested controller ensures precise trajectory tracking in the presence of slip.
Dual-Space Adaptive Control of Redundantly Actuated Parallel Manipulators for Extremely Fast Operations with Load ChangesThis paper deals with the dual-space adaptive control of R4 redundantly actuated parallel manipulator for applications with very high accelerations. This controller is compared experimentally with a dual-space feedforward controller (which may have good performances for specific cases, but has crucial losses of performance when there is any operational change (such as a change of load)), for a pick-and-place task with accelerations of 30G (without payload)and 20G (with a payload of 200g). The objective of this paper is to show that the proposed dual-space adaptive controller not only keeps a very good performance independently of the operational case, but also has a better performance than the dual-space feedforward controller even when this last one is best configured to the given case.
Learning Tracking Control with Forward ModelsPerforming task-space tracking control on redundant robot manipulators is a difficult problem. When the physical model of the robot is too complex or not available, standard methods fail and machine learning algorithms can have advantages. We propose an adaptive learning algorithm for tracking control of underactuated or non-rigid robots where the physical model of the robot is unavailable. The control method is based on the fact that forward models are relatively straightforward to learn and local inversions can be obtained via local optimization. We use sparse online Gaussian process inference to obtain a flexible probabilistic forward model and second order optimization to find the inverse mapping. Physical experiments indicate that this approach can outperform state-of-the-art tracking control algorithms in this context.
Predictive Gaze Stabilization During Periodic Locomotion Based on Adaptive Frequency OscillatorsIn this paper we present an approach to the problem of stabilizating the gaze of legged robots using Adaptive Frequency Oscillators to learn the frequency, phase and amplitude of the optical flow and generate compensatory commands during robot locomotion. Assuming periodic and nearly sine shaped motion of the head of the robot, the system successfully stabilizes the gaze of the robot, whether the robot itself is moving, or an external object is moving relative to the robot. We present experiments in simulation and, for object tracking, with a real robotics setup, the Hoap 3, showing that the system can be successfully applied to gaze stabilization during locomotion, even when the feedback loop is very slow and noisy.
Learning-Based Model Predictive Control on a Quadrotor: Onboard Implementation and Experimental ResultsIn this paper, we present details of the real time implementation onboard a quadrotor helicopter of learning-based model predictive control (LBMPC). LBMPC rigorously combines statistical learning with control engineering, while providing levels of guarantees about safety, robustness, and convergence. Experimental results show that LBMPC can learn physically based updates to an initial model, and how as a result LBMPC improves transient response performance. We demonstrate robustness to mis-learning. Finally, we show the use of LBMPC in an integrated robotic task demonstration---The quadrotor is used to catch a ball thrown with an a priori unknown trajectory.