Header logo is


2017


Multi-Modal Imitation Learning from Unstructured Demonstrations using Generative Adversarial Nets
Multi-Modal Imitation Learning from Unstructured Demonstrations using Generative Adversarial Nets

Hausman, K., Chebotar, Y., Schaal, S., Sukhatme, G., Lim, J.

In Proceedings from the conference "Neural Information Processing Systems 2017., (Editors: Guyon I. and Luxburg U.v. and Bengio S. and Wallach H. and Fergus R. and Vishwanathan S. and Garnett R.), Curran Associates, Inc., Advances in Neural Information Processing Systems 30 (NIPS), December 2017 (inproceedings)

am

pdf video [BibTex]

2017


pdf video [BibTex]


On the Design of {LQR} Kernels for Efficient Controller Learning
On the Design of LQR Kernels for Efficient Controller Learning

Marco, A., Hennig, P., Schaal, S., Trimpe, S.

Proceedings of the 56th IEEE Annual Conference on Decision and Control (CDC), pages: 5193-5200, IEEE, IEEE Conference on Decision and Control, December 2017 (conference)

Abstract
Finding optimal feedback controllers for nonlinear dynamic systems from data is hard. Recently, Bayesian optimization (BO) has been proposed as a powerful framework for direct controller tuning from experimental trials. For selecting the next query point and finding the global optimum, BO relies on a probabilistic description of the latent objective function, typically a Gaussian process (GP). As is shown herein, GPs with a common kernel choice can, however, lead to poor learning outcomes on standard quadratic control problems. For a first-order system, we construct two kernels that specifically leverage the structure of the well-known Linear Quadratic Regulator (LQR), yet retain the flexibility of Bayesian nonparametric learning. Simulations of uncertain linear and nonlinear systems demonstrate that the LQR kernels yield superior learning performance.

am ics pn

arXiv PDF On the Design of LQR Kernels for Efficient Controller Learning - CDC presentation DOI Project Page [BibTex]

arXiv PDF On the Design of LQR Kernels for Efficient Controller Learning - CDC presentation DOI Project Page [BibTex]


Optimizing Long-term Predictions for Model-based Policy Search
Optimizing Long-term Predictions for Model-based Policy Search

Doerr, A., Daniel, C., Nguyen-Tuong, D., Marco, A., Schaal, S., Toussaint, M., Trimpe, S.

Proceedings of 1st Annual Conference on Robot Learning (CoRL), 78, pages: 227-238, (Editors: Sergey Levine and Vincent Vanhoucke and Ken Goldberg), 1st Annual Conference on Robot Learning, November 2017 (conference)

Abstract
We propose a novel long-term optimization criterion to improve the robustness of model-based reinforcement learning in real-world scenarios. Learning a dynamics model to derive a solution promises much greater data-efficiency and reusability compared to model-free alternatives. In practice, however, modelbased RL suffers from various imperfections such as noisy input and output data, delays and unmeasured (latent) states. To achieve higher resilience against such effects, we propose to optimize a generative long-term prediction model directly with respect to the likelihood of observed trajectories as opposed to the common approach of optimizing a dynamics model for one-step-ahead predictions. We evaluate the proposed method on several artificial and real-world benchmark problems and compare it to PILCO, a model-based RL framework, in experiments on a manipulation robot. The results show that the proposed method is competitive compared to state-of-the-art model learning methods. In contrast to these more involved models, our model can directly be employed for policy search and outperforms a baseline method in the robot experiment.

am ics

PDF Project Page [BibTex]

PDF Project Page [BibTex]


no image
A New Data Source for Inverse Dynamics Learning

Kappler, D., Meier, F., Ratliff, N., Schaal, S.

In Proceedings IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), IEEE, Piscataway, NJ, USA, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), September 2017 (inproceedings)

am

[BibTex]

[BibTex]


no image
Bayesian Regression for Artifact Correction in Electroencephalography

Fiebig, K., Jayaram, V., Hesse, T., Blank, A., Peters, J., Grosse-Wentrup, M.

Proceedings of the 7th Graz Brain-Computer Interface Conference 2017 - From Vision to Reality, pages: 131-136, (Editors: Müller-Putz G.R., Steyrl D., Wriessnegger S. C., Scherer R.), Graz University of Technology, Austria, Graz Brain-Computer Interface Conference, September 2017 (conference)

am ei

DOI [BibTex]

DOI [BibTex]


no image
Investigating Music Imagery as a Cognitive Paradigm for Low-Cost Brain-Computer Interfaces

Grossberger, L., Hohmann, M. R., Peters, J., Grosse-Wentrup, M.

Proceedings of the 7th Graz Brain-Computer Interface Conference 2017 - From Vision to Reality, pages: 160-164, (Editors: Müller-Putz G.R., Steyrl D., Wriessnegger S. C., Scherer R.), Graz University of Technology, Austria, Graz Brain-Computer Interface Conference, September 2017 (conference)

am ei

DOI [BibTex]

DOI [BibTex]


On the relevance of grasp metrics for predicting grasp success
On the relevance of grasp metrics for predicting grasp success

Rubert, C., Kappler, D., Morales, A., Schaal, S., Bohg, J.

In Proceedings of the IEEE/RSJ International Conference of Intelligent Robots and Systems, September 2017 (inproceedings) Accepted

Abstract
We aim to reliably predict whether a grasp on a known object is successful before it is executed in the real world. There is an entire suite of grasp metrics that has already been developed which rely on precisely known contact points between object and hand. However, it remains unclear whether and how they may be combined into a general purpose grasp stability predictor. In this paper, we analyze these questions by leveraging a large scale database of simulated grasps on a wide variety of objects. For each grasp, we compute the value of seven metrics. Each grasp is annotated by human subjects with ground truth stability labels. Given this data set, we train several classification methods to find out whether there is some underlying, non-trivial structure in the data that is difficult to model manually but can be learned. Quantitative and qualitative results show the complexity of the prediction problem. We found that a good prediction performance critically depends on using a combination of metrics as input features. Furthermore, non-parametric and non-linear classifiers best capture the structure in the data.

am

Project Page [BibTex]

Project Page [BibTex]


no image
Swimming in low reynolds numbers using planar and helical flagellar waves

Khalil, I. S. M., Tabak, A. F., Seif, M. A., Klingner, A., Adel, B., Sitti, M.

In International Conference on Intelligent Robots and Systems (IROS) 2017, pages: 1907-1912, International Conference on Intelligent Robots and Systems, September 2017 (inproceedings)

Abstract
In travelling towards the oviducts, sperm cells undergo transitions between planar to helical flagellar propulsion by a beating tail based on the viscosity of the environment. In this work, we aim to model and mimic this behaviour in low Reynolds number fluids using externally actuated soft robotic sperms. We numerically investigate the effects of transition between planar to helical flagellar propulsion on the swimming characteristics of the robotic sperm using a model based on resistive-force theory to study the role of viscous forces on its flexible tail. Experimental results are obtained using robots that contain magnetic particles within the polymer matrix of its head and an ultra-thin flexible tail. The planar and helical flagellar propulsion are achieved using in-plane and out-of-plane uniform fields with sinusoidally varying components, respectively. We experimentally show that the swimming speed of the robotic sperm increases by a factor of 1.4 (fluid viscosity 5 Pa.s) when it undergoes a controlled transition between planar to helical flagellar propulsion, at relatively low actuation frequencies.

pi

DOI [BibTex]

DOI [BibTex]


no image
Local Bayesian Optimization of Motor Skills

Akrour, R., Sorokin, D., Peters, J., Neumann, G.

Proceedings of the 34th International Conference on Machine Learning, 70, pages: 41-50, Proceedings of Machine Learning Research, (Editors: Doina Precup, Yee Whye Teh), PMLR, International Conference on Machine Learning (ICML), August 2017 (conference)

am ei

link (url) Project Page [BibTex]

link (url) Project Page [BibTex]


Combining Model-Based and Model-Free Updates for Trajectory-Centric Reinforcement Learning
Combining Model-Based and Model-Free Updates for Trajectory-Centric Reinforcement Learning

Chebotar, Y., Hausman, K., Zhang, M., Sukhatme, G., Schaal, S., Levine, S.

Proceedings of the 34th International Conference on Machine Learning, 70, Proceedings of Machine Learning Research, (Editors: Doina Precup, Yee Whye Teh), PMLR, International Conference on Machine Learning (ICML), August 2017 (conference)

am

pdf video [BibTex]

pdf video [BibTex]


An XY ϴz flexure mechanism with optimal stiffness properties
An XY ϴz flexure mechanism with optimal stiffness properties

Lum, G. Z., Pham, M. T., Teo, T. J., Yang, G., Yeo, S. H., Sitti, M.

In 2017 IEEE International Conference on Advanced Intelligent Mechatronics (AIM), pages: 1103-1110, July 2017 (inproceedings)

Abstract
The development of optimal XY θz flexure mechanisms, which can deliver high precision motion about the z-axis, and along the x- and y-axes is highly desirable for a wide range of micro/nano-positioning tasks pertaining to biomedical research, microscopy technologies and various industrial applications. Although maximizing the stiffness ratios is a very critical design requirement, the achievable translational and rotational stiffness ratios of existing XY θz flexure mechanisms are still restricted between 0.5 and 130. As a result, these XY θz flexure mechanisms are unable to fully optimize their workspace and capabilities to reject disturbances. Here, we present an optimal XY θz flexure mechanism, which is designed to have maximum stiffness ratios. Based on finite element analysis (FEA), it has translational stiffness ratio of 248, rotational stiffness ratio of 238 and a large workspace of 2.50 mm × 2.50 mm × 10°. Despite having such a large workspace, FEA also predicts that the proposed mechanism can still achieve a high bandwidth of 70 Hz. In comparison, the bandwidth of similar existing flexure mechanisms that can deflect more than 0.5 mm or 0.5° is typically less than 45 Hz. Hence, the high stiffness ratios of the proposed mechanism are achieved without compromising its dynamic performance. Preliminary experimental results pertaining to the mechanism's translational actuating stiffness and bandwidth were in agreement with the FEA predictions as the deviation was within 10%. In conclusion, the proposed flexure mechanism exhibits superior performance and can be used across a wide range of applications.

pi

DOI [BibTex]

DOI [BibTex]


Positioning of drug carriers using permanent magnet-based robotic system in three-dimensional space
Positioning of drug carriers using permanent magnet-based robotic system in three-dimensional space

Khalil, I. S. M., Alfar, A., Tabak, A. F., Klingner, A., Stramigioli, S., Sitti, M.

In 2017 IEEE International Conference on Advanced Intelligent Mechatronics (AIM), pages: 1117-1122, July 2017 (inproceedings)

Abstract
Magnetic control of drug carriers using systems with open-configurations is essential to enable scaling to the size of in vivo applications. In this study, we demonstrate motion control of paramagnetic microparticles in a low Reynolds number fluid, using a permanent magnet-based robotic system with an open-configuration. The microparticles are controlled in three-dimensional (3D) space using a cylindrical NdFeB magnet that is fixed to the end-effector of a robotic arm. We develop a kinematic map between the position of the microparticles and the configuration of the robotic arm, and use this map as a basis of a closed-loop control system based on the position of the microparticles. Our experimental results show the ability of the robot configuration to control the exerted field gradient on the dipole of the microparticles, and achieve positioning in 3D space with maximum error of 300 µm and 600 µm in the steady-state during setpoint and trajectory tracking, respectively.

pi

DOI [BibTex]

DOI [BibTex]


no image
Self-assembly of micro/nanosystems across scales and interfaces

Mastrangeli, M.

In 2017 19th International Conference on Solid-State Sensors, Actuators and Microsystems (TRANSDUCERS), pages: 676 - 681, IEEE, July 2017 (inproceedings)

Abstract
Steady progress in understanding and implementation are establishing self-assembly as a versatile, parallel and scalable approach to the fabrication of transducers. In this contribution, I illustrate the principles and reach of self-assembly with three applications at different scales - namely, the capillary self-alignment of millimetric components, the sealing of liquid-filled polymeric microcapsules, and the accurate capillary assembly of single nanoparticles - and propose foreseeable directions for further developments.

pi

link (url) DOI [BibTex]

link (url) DOI [BibTex]


Dynamic analysis on hexapedal water-running robot with compliant joints
Dynamic analysis on hexapedal water-running robot with compliant joints

Kim, H., Liu, Y., Jeong, K., Sitti, M., Seo, T.

In 2017 14th International Conference on Ubiquitous Robots and Ambient Intelligence (URAI), pages: 250-251, June 2017 (inproceedings)

Abstract
The dynamic analysis has been considered as one of the important design methods to design robots. In this research, we derive dynamic equation of hexapedal water-running robot to design compliant joints. The compliant joints that connect three bodies will be used to improve mobility and stability of water-running motion's pitch behavior. We considered all of parts as rigid body including links of six Klann mechanisms and three main frames. And then, we derived dynamic equation by using the Lagrangian method with external force of the water. We are expecting that the dynamic analysis is going to be used to design parts of the water running robot.

pi

DOI [BibTex]

DOI [BibTex]


Model-Based Policy Search for Automatic Tuning of Multivariate PID Controllers
Model-Based Policy Search for Automatic Tuning of Multivariate PID Controllers

Doerr, A., Nguyen-Tuong, D., Marco, A., Schaal, S., Trimpe, S.

In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pages: 5295-5301, IEEE, Piscataway, NJ, USA, IEEE International Conference on Robotics and Automation (ICRA), May 2017 (inproceedings)

am ics

PDF arXiv DOI Project Page [BibTex]

PDF arXiv DOI Project Page [BibTex]


Learning Feedback Terms for Reactive Planning and Control
Learning Feedback Terms for Reactive Planning and Control

Rai, A., Sutanto, G., Schaal, S., Meier, F.

Proceedings 2017 IEEE International Conference on Robotics and Automation (ICRA), IEEE, Piscataway, NJ, USA, IEEE International Conference on Robotics and Automation (ICRA), May 2017 (conference)

am

pdf video [BibTex]

pdf video [BibTex]


Design and actuation of a magnetic millirobot under a constant unidirectional magnetic field
Design and actuation of a magnetic millirobot under a constant unidirectional magnetic field

Erin, O., Giltinan, J., Tsai, L., Sitti, M.

In Proceedings 2017 IEEE International Conference on Robotics and Automation (ICRA), pages: 3404-3410, IEEE, Piscataway, NJ, USA, IEEE International Conference on Robotics and Automation (ICRA), May 2017 (inproceedings)

Abstract
Magnetic untethered millirobots, which are actuated and controlled by remote magnetic fields, have been proposed for medical applications due to their ability to safely pass through tissues at long ranges. For example, magnetic resonance imaging (MRI) systems with a 3-7 T constant unidirectional magnetic field and 3D gradient coils have been used to actuate magnetic robots. Such magnetically constrained systems place limits on the degrees of freedom that can be actuated for untethered devices. This paper presents a design and actuation methodology for a magnetic millirobot that exhibits both position and orientation control in 2D under a magnetic field, dominated by a constant unidirectional magnetic field as found in MRI systems. Placing a spherical permanent magnet, which is free to rotate inside the millirobot and located away from the center of mass, allows the generation of net forces and torques with applied 3D magnetic field gradients. We model this system in a 3D planar case and experimentally demonstrate open-loop control of both position and orientation by the applied 2D field gradients. The actuation performance is characterized across the most important design variables, and we experimentally demonstrate that the proposed approach is feasible.

pi

DOI [BibTex]

DOI [BibTex]


Virtual vs. {R}eal: Trading Off Simulations and Physical Experiments in Reinforcement Learning with {B}ayesian Optimization
Virtual vs. Real: Trading Off Simulations and Physical Experiments in Reinforcement Learning with Bayesian Optimization

Marco, A., Berkenkamp, F., Hennig, P., Schoellig, A. P., Krause, A., Schaal, S., Trimpe, S.

In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pages: 1557-1563, IEEE, Piscataway, NJ, USA, IEEE International Conference on Robotics and Automation (ICRA), May 2017 (inproceedings)

am ics pn

PDF arXiv ICRA 2017 Spotlight presentation Virtual vs. Real - Video explanation DOI Project Page [BibTex]

PDF arXiv ICRA 2017 Spotlight presentation Virtual vs. Real - Video explanation DOI Project Page [BibTex]


Magnetically actuated soft capsule endoscope for fine-needle aspiration biopsy
Magnetically actuated soft capsule endoscope for fine-needle aspiration biopsy

Son, D., Dogan, M. D., Sitti, M.

In Proceedings 2017 IEEE International Conference on Robotics and Automation (ICRA), pages: 1132-1139, IEEE, Piscataway, NJ, USA, IEEE International Conference on Robotics and Automation (ICRA), May 2017 (inproceedings)

Abstract
This paper presents a magnetically actuated soft capsule endoscope for fine-needle aspiration biopsy (B-MASCE) in the upper gastrointestinal tract. A thin and hollow needle is attached to the capsule, which can penetrate deeply into tissues to obtain subsurface biopsy sample. The design utilizes a soft elastomer body as a compliant mechanism to guide the needle. An internal permanent magnet provides a means for both actuation and tracking. The capsule is designed to roll towards its target and then deploy the biopsy needle in a precise location selected as the target area. B-MASCE is controlled by multiple custom-designed electromagnets while its position and orientation are tracked by a magnetic sensor array. In in vitro trials, B-MASCE demonstrated rolling locomotion and biopsy of a swine tissue model positioned inside an anatomical human stomach model. It was confirmed after the experiment that a tissue sample was retained inside the needle.

pi

DOI Project Page [BibTex]

DOI Project Page [BibTex]


The use of clamping grips and friction pads by tree frogs for climbing curved surfaces
The use of clamping grips and friction pads by tree frogs for climbing curved surfaces

Endlein, T., Ji, A., Yuan, S., Hill, I., Wang, H., Barnes, W. J. P., Dai, Z., Sitti, M.

In Proc. R. Soc. B, 284(1849):20162867, Febuary 2017 (inproceedings)

Abstract
Most studies on the adhesive mechanisms of climbing animals have addressed attachment against flat surfaces, yet many animals can climb highly curved surfaces, like twigs and small branches. Here we investigated whether tree frogs use a clamping grip by recording the ground reaction forces on a cylindrical object with either a smooth or anti-adhesive, rough surface. Furthermore, we measured the contact area of fore and hindlimbs against differently sized transparent cylinders and the forces of individual pads and subarticular tubercles in restrained animals. Our study revealed that frogs use friction and normal forces of roughly a similar magnitude for holding on to cylindrical objects. When challenged with climbing a non-adhesive surface, the compressive forces between opposite legs nearly doubled, indicating a stronger clamping grip. In contrast to climbing flat surfaces, frogs increased the contact area on all limbs by engaging not just adhesive pads but also subarticular tubercles on curved surfaces. Our force measurements showed that tubercles can withstand larger shear stresses than pads. SEM images of tubercles revealed a similar structure to that of toe pads including the presence of nanopillars, though channels surrounding epithelial cells were less pronounced. The tubercles' smaller size, proximal location on the toes and shallow cells make them probably less prone to buckling and thus ideal for gripping curved surfaces.

pi

DOI [BibTex]

DOI [BibTex]


no image
Robot Learning

Peters, J., Lee, D., Kober, J., Nguyen-Tuong, D., Bagnell, J., Schaal, S.

In Springer Handbook of Robotics, pages: 357-394, 15, 2nd, (Editors: Siciliano, Bruno and Khatib, Oussama), Springer International Publishing, 2017 (inbook)

am ei

Project Page [BibTex]

Project Page [BibTex]


Planning spin-walking locomotion for automatic grasping of microobjects by an untethered magnetic microgripper
Planning spin-walking locomotion for automatic grasping of microobjects by an untethered magnetic microgripper

Dong, X., Sitti, M.

In 2017 IEEE International Conference on Robotics and Automation (ICRA), pages: 6612-6618, 2017 (inproceedings)

Abstract
Most demonstrated mobile microrobot tasks so far have been achieved via pick-and-placing and dynamic trapping with teleoperation or simple path following algorithms. In our previous work, an untethered magnetic microgripper has been developed which has advanced functions, such as gripping objects. Both teleoperated manipulation in 2D and 3D have been demonstrated. However, it is challenging to control the magnetic microgripper to carry out manipulation tasks, because the grasping of objects so far in the literature relies heavily on teleoperation, which takes several minutes with even a skilled human expert. Here, we propose a new spin-walking locomotion and an automated 2D grasping motion planner for the microgripper, which enables time-efficient automatic grasping of microobjects that has not been achieved yet for untethered microrobots. In its locomotion, the microgripper repeatedly rotates about two principal axes to regulate its pose and move precisely on a surface. The motion planner could plan different motion primitives for grasping and compensate the uncertainties in the motion by learning the uncertainties and planning accordingly. We experimentally demonstrated that, using the proposed method, the microgripper could align to the target pose with error less than 0.1 body length and grip the objects within 40 seconds. Our method could significantly improve the time efficiency of micro-scale manipulation and have potential applications in microassembly and biomedical engineering.

pi

DOI Project Page [BibTex]

DOI Project Page [BibTex]


Methods, apparatuses, and systems for micromanipulation with adhesive fibrillar structures
Methods, apparatuses, and systems for micromanipulation with adhesive fibrillar structures

Sitti, M., Mengüç, Y.

US Patent 9,731,422, 2017 (patent)

Abstract
The present invention are methods for fabrication of micro- and/or nano-scale adhesive fibers and their use for movement and manipulation of objects. Further disclosed is a method of manipulating a part by providing a manipulation device with a plurality of fibers, where each fiber has a tip with a flat surface that is parallel to a backing layer, contacting the flat surfaces on an object, moving the object to a new location, then disengaging the tips from the object.

pi

link (url) [BibTex]

2010


no image
Reinforcement learning of full-body humanoid motor skills

Stulp, F., Buchli, J., Theodorou, E., Schaal, S.

In Humanoid Robots (Humanoids), 2010 10th IEEE-RAS International Conference on, pages: 405-410, December 2010, clmc (inproceedings)

Abstract
Applying reinforcement learning to humanoid robots is challenging because humanoids have a large number of degrees of freedom and state and action spaces are continuous. Thus, most reinforcement learning algorithms would become computationally infeasible and require a prohibitive amount of trials to explore such high-dimensional spaces. In this paper, we present a probabilistic reinforcement learning approach, which is derived from the framework of stochastic optimal control and path integrals. The algorithm, called Policy Improvement with Path Integrals (PI2), has a surprisingly simple form, has no open tuning parameters besides the exploration noise, is model-free, and performs numerically robustly in high dimensional learning problems. We demonstrate how PI2 is able to learn full-body motor skills on a 34-DOF humanoid robot. To demonstrate the generality of our approach, we also apply PI2 in the context of variable impedance control, where both planned trajectories and gain schedules for each joint are optimized simultaneously.

am

link (url) [BibTex]

2010


link (url) [BibTex]


no image
Relative Entropy Policy Search

Peters, J., Mülling, K., Altun, Y.

In Proceedings of the Twenty-Fourth National Conference on Artificial Intelligence, pages: 1607-1612, (Editors: Fox, M. , D. Poole), AAAI Press, Menlo Park, CA, USA, Twenty-Fourth National Conference on Artificial Intelligence (AAAI-10), July 2010 (inproceedings)

Abstract
Policy search is a successful approach to reinforcement learning. However, policy improvements often result in the loss of information. Hence, it has been marred by premature convergence and implausible solutions. As first suggested in the context of covariant policy gradients (Bagnell and Schneider 2003), many of these problems may be addressed by constraining the information loss. In this paper, we continue this path of reasoning and suggest the Relative Entropy Policy Search (REPS) method. The resulting method differs significantly from previous policy gradient approaches and yields an exact update step. It works well on typical reinforcement learning benchmark problems.

am ei

PDF Web [BibTex]

PDF Web [BibTex]


no image
Reinforcement learning of motor skills in high dimensions: A path integral approach

Theodorou, E., Buchli, J., Schaal, S.

In Robotics and Automation (ICRA), 2010 IEEE International Conference on, pages: 2397-2403, May 2010, clmc (inproceedings)

Abstract
Reinforcement learning (RL) is one of the most general approaches to learning control. Its applicability to complex motor systems, however, has been largely impossible so far due to the computational difficulties that reinforcement learning encounters in high dimensional continuous state-action spaces. In this paper, we derive a novel approach to RL for parameterized control policies based on the framework of stochastic optimal control with path integrals. While solidly grounded in optimal control theory and estimation theory, the update equations for learning are surprisingly simple and have no danger of numerical instabilities as neither matrix inversions nor gradient learning rates are required. Empirical evaluations demonstrate significant performance improvements over gradient-based policy learning and scalability to high-dimensional control problems. Finally, a learning experiment on a robot dog illustrates the functionality of our algorithm in a real-world scenario. We believe that our new algorithm, Policy Improvement with Path Integrals (PI2), offers currently one of the most efficient, numerically robust, and easy to implement algorithms for RL in robotics.

am

link (url) [BibTex]

link (url) [BibTex]


no image
Inverse dynamics control of floating base systems using orthogonal decomposition

Mistry, M., Buchli, J., Schaal, S.

In Robotics and Automation (ICRA), 2010 IEEE International Conference on, pages: 3406-3412, May 2010, clmc (inproceedings)

Abstract
Model-based control methods can be used to enable fast, dexterous, and compliant motion of robots without sacrificing control accuracy. However, implementing such techniques on floating base robots, e.g., humanoids and legged systems, is non-trivial due to under-actuation, dynamically changing constraints from the environment, and potentially closed loop kinematics. In this paper, we show how to compute the analytically correct inverse dynamics torques for model-based control of sufficiently constrained floating base rigid-body systems, such as humanoid robots with one or two feet in contact with the environment. While our previous inverse dynamics approach relied on an estimation of contact forces to compute an approximate inverse dynamics solution, here we present an analytically correct solution by using an orthogonal decomposition to project the robot dynamics onto a reduced dimensional space, independent of contact forces. We demonstrate the feasibility and robustness of our approach on a simulated floating base bipedal humanoid robot and an actual robot dog locomoting over rough terrain.

am

link (url) [BibTex]

link (url) [BibTex]


no image
Fast, robust quadruped locomotion over challenging terrain

Kalakrishnan, M., Buchli, J., Pastor, P., Mistry, M., Schaal, S.

In Robotics and Automation (ICRA), 2010 IEEE International Conference on, pages: 2665-2670, May 2010, clmc (inproceedings)

Abstract
We present a control architecture for fast quadruped locomotion over rough terrain. We approach the problem by decomposing it into many sub-systems, in which we apply state-of-the-art learning, planning, optimization and control techniques to achieve robust, fast locomotion. Unique features of our control strategy include: (1) a system that learns optimal foothold choices from expert demonstration using terrain templates, (2) a body trajectory optimizer based on the Zero-Moment Point (ZMP) stability criterion, and (3) a floating-base inverse dynamics controller that, in conjunction with force control, allows for robust, compliant locomotion over unperceived obstacles. We evaluate the performance of our controller by testing it on the LittleDog quadruped robot, over a wide variety of rough terrain of varying difficulty levels. We demonstrate the generalization ability of this controller by presenting test results from an independent external test team on terrains that have never been shown to us.

am

link (url) [BibTex]

link (url) [BibTex]


no image
Adhesion recovery and passive peeling in a wall climbing robot using adhesives

Kute, C., Murphy, M. P., Mengüç, Y., Sitti, M.

In Robotics and Automation (ICRA), 2010 IEEE International Conference on, pages: 2797-2802, 2010 (inproceedings)

pi

[BibTex]

[BibTex]


no image
Nanohandling robot cells

Fatikow, Sergej, Wich, Thomas, Dahmen, Christian, Jasper, Daniel, Stolle, Christian, Eichhorn, Volkmar, Hagemann, Saskia, Weigel-Jech, Michael

In Handbook of Nanophysics: Nanomedicine and Nanorobotics, pages: 1-31, CRC Press, 2010 (incollection)

pi

[BibTex]

[BibTex]


no image
Locally weighted regression for control

Ting, J., Vijayakumar, S., Schaal, S.

In Encyclopedia of Machine Learning, pages: 613-624, (Editors: Sammut, C.;Webb, G. I.), Springer, 2010, clmc (inbook)

Abstract
This is article addresses two topics: learning control and locally weighted regression.

am

link (url) [BibTex]

link (url) [BibTex]


no image
Are reaching movements planned in kinematic or dynamic coordinates?

Ellmer, A., Schaal, S.

In Abstracts of Neural Control of Movement Conference (NCM 2010), Naples, Florida, 2010, 2010, clmc (inproceedings)

Abstract
Whether human reaching movements are planned and optimized in kinematic (task space) or dynamic (joint or muscle space) coordinates is still an issue of debate. The first hypothesis implies that a planner produces a desired end-effector position at each point in time during the reaching movement, whereas the latter hypothesis includes the dynamics of the muscular-skeletal control system to produce a continuous end-effector trajectory. Previous work by Wolpert et al (1995) showed that when subjects were led to believe that their straight reaching paths corresponded to curved paths as shown on a computer screen, participants adapted the true path of their hand such that they would visually perceive a straight line in visual space, despite that they actually produced a curved path. These results were interpreted as supporting the stance that reaching trajectories are planned in kinematic coordinates. However, this experiment could only demonstrate that adaptation to altered paths, i.e. the position of the end-effector, did occur, but not that the precise timing of end-effector position was equally planned, i.e., the trajectory. Our current experiment aims at filling this gap by explicitly testing whether position over time, i.e. velocity, is a property of reaching movements that is planned in kinematic coordinates. In the current experiment, the velocity profiles of cursor movements corresponding to the participant's hand motions were skewed either to the left or to the right; the path itself was left unaltered. We developed an adaptation paradigm, where the skew of the velocity profile was introduced gradually and participants reported no awareness of any manipulation. Preliminary results indicate that the true hand motion of participants did not alter, i.e. there was no adaptation so as to counterbalance the introduced skew. However, for some participants, peak hand velocities were lowered for higher skews, which suggests that participants interpreted the manipulation as mere noise due to variance in their own movement. In summary, for a visuomotor transformation task, the hypothesis of a planned continuous end-effector trajectory predicts adaptation to a modified velocity profile. The current experiment found no systematic adaptation under such transformation, but did demonstrate an effect that is more in accordance that subjects could not perceive the manipulation and rather interpreted as an increase of noise.

am

[BibTex]

[BibTex]


no image
Comparison of linear and nonlinear buck converter models with varying compensator gain values for design optimization

Sattler, Michael, Lui, Yusi, Edrington, Chris S

In North American Power Symposium (NAPS), 2010, pages: 1-7, 2010 (inproceedings)

pi

[BibTex]

[BibTex]


no image
Optimality in Neuromuscular Systems

Theodorou, E. A., Valero-Cuevas, F.

In 32nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2010, clmc (inproceedings)

Abstract
Abstract? We provide an overview of optimal control meth- ods to nonlinear neuromuscular systems and discuss their lim- itations. Moreover we extend current optimal control methods to their application to neuromuscular models with realistically numerous musculotendons; as most prior work is limited to torque-driven systems. Recent work on computational motor control has explored the used of control theory and esti- mation as a conceptual tool to understand the underlying computational principles of neuromuscular systems. After all, successful biological systems regularly meet conditions for stability, robustness and performance for multiple classes of complex tasks. Among a variety of proposed control theory frameworks to explain this, stochastic optimal control has become a dominant framework to the point of being a standard computational technique to reproduce kinematic trajectories of reaching movements (see [12]) In particular, we demonstrate the application of optimal control to a neuromuscular model of the index finger with all seven musculotendons producing a tapping task. Our simu- lations include 1) a muscle model that includes force- length and force-velocity characteristics; 2) an anatomically plausible biomechanical model of the index finger that includes a tendi- nous network for the extensor mechanism and 3) a contact model that is based on a nonlinear spring-damper attached at the end effector of the index finger. We demonstrate that it is feasible to apply optimal control to systems with realistically large state vectors and conclude that, while optimal control is an adequate formalism to create computational models of neuro- musculoskeletal systems, there remain important challenges and limitations that need to be considered and overcome such as contact transitions, curse of dimensionality, and constraints on states and controls.

am

PDF [BibTex]

PDF [BibTex]


no image
Enhancing the performance of Bio-inspired adhesives

Chung, H., Glass, P., Sitti, M., Washburn, N. R.

In ABSTRACTS OF PAPERS OF THE AMERICAN CHEMICAL SOCIETY, 240, 2010 (inproceedings)

pi

[BibTex]

[BibTex]


no image
Control performance simulation in the design of a flapping wing micro-aerial vehicle

Hines, L. L., Arabagi, V., Sitti, M.

In Intelligent Robots and Systems (IROS), 2010 IEEE/RSJ International Conference on, pages: 1090-1095, 2010 (inproceedings)

pi

Project Page [BibTex]

Project Page [BibTex]


no image
Learning Policy Improvements with Path Integrals

Theodorou, E. A., Buchli, J., Schaal, S.

In International Conference on Artificial Intelligence and Statistics (AISTATS 2010), 2010, clmc (inproceedings)

Abstract
With the goal to generate more scalable algo- rithms with higher efficiency and fewer open parameters, reinforcement learning (RL) has recently moved towards combining classi- cal techniques from optimal control and dy- namic programming with modern learning techniques from statistical estimation the- ory. In this vein, this paper suggests the framework of stochastic optimal control with path integrals to derive a novel approach to RL with parametrized policies. While solidly grounded in value function estimation and optimal control based on the stochastic Hamilton-Jacobi-Bellman (HJB) equations, policy improvements can be transformed into an approximation problem of a path inte- gral which has no open parameters other than the exploration noise. The resulting algorithm can be conceived of as model- based, semi-model-based, or even model free, depending on how the learning problem is structured. Our new algorithm demon- strates interesting similarities with previous RL research in the framework of proba- bility matching and provides intuition why the slightly heuristically motivated proba- bility matching approach can actually per- form well. Empirical evaluations demon- strate significant performance improvements over gradient-based policy learning and scal- ability to high-dimensional control problems. We believe that Policy Improvement with Path Integrals (PI2) offers currently one of the most efficient, numerically robust, and easy to implement algorithms for RL based on trajectory roll-outs.

am

PDF [BibTex]

PDF [BibTex]


no image
Learning optimal control solutions: a path integral approach

Theodorou, E., Schaal, S.

In Abstracts of Neural Control of Movement Conference (NCM 2010), Naples, Florida, 2010, 2010, clmc (inproceedings)

Abstract
Investigating principles of human motor control in the framework of optimal control has had a long tradition in neural control of movement, and has recently experienced a new surge of investigations. Ideally, optimal control problems are addresses as a reinforcement learning (RL) problem, which would allow to investigate both the process of acquiring an optimal control solution as well as the solution itself. Unfortunately, the applicability of RL to complex neural and biomechanics systems has been largely impossible so far due to the computational difficulties that arise in high dimensional continuous state-action spaces. As a way out, research has focussed on computing optimal control solutions based on iterative optimal control methods that are based on linear and quadratic approximations of dynamical models and cost functions. These methods require perfect knowledge of the dynamics and cost functions while they are based on gradient and Newton optimization schemes. Their applicability is also restricted to low dimensional problems due to problematic convergence in high dimensions. Moreover, the process of computing the optimal solution is removed from the learning process that might be plausible in biology. In this work, we present a new reinforcement learning method for learning optimal control solutions or motor control. This method, based on the framework of stochastic optimal control with path integrals, has a very solid theoretical foundation, while resulting in surprisingly simple learning algorithms. It is also possible to apply this approach without knowledge of the system model, and to use a wide variety of complex nonlinear cost functions for optimization. We illustrate the theoretical properties of this approach and its applicability to learning motor control tasks for reaching movements and locomotion studies. We discuss its applicability to learning desired trajectories, variable stiffness control (co-contraction), and parameterized control policies. We also investigate the applicability to signal dependent noise control systems. We believe that the suggested method offers one of the easiest to use approaches to learning optimal control suggested in the literature so far, which makes it ideally suited for computational investigations of biological motor control.

am

[BibTex]

[BibTex]


no image
A highly integrated low pressure fluid servo-valve for applications in wearable robotic systems

Folgheraiter, M., Jordan, M., Vaca Benitez, L. M., Grimminger, F., Schmidt, S., Albiez, J., Kirchner, F.

In ICINCO 2010 - Proceedings of the 7th International Conference on Informatics in Control, Automation and Robotics, 2, pages: 72-78, 2010 (inproceedings)

Abstract
In this paper an innovative low pressure servo-valve is presented. The device was designed with the main aim to be easily integrable into complex hydraulic/pneumatic actuation systems, and to operate at relatively low pressure (<50·105Pa). Characteristics like compactness, lightweight, high bandwidth, and autonomous sensory capability, where considered during the design process in order to achieve a device that fulfills the basic requirements for a wearable robotic system. Preliminary results about the prototype performances are presented here, in particular its dynamic behavior was measured for different working conditions, and a non-linear model identified using a recursive Hammerstein-Wiener parameter adaptation algorithm.

am

[BibTex]

[BibTex]


no image
Constrained Accelerations for Controlled Geometric Reduction: Sagittal-Plane Decoupling for Bipedal Locomotion

Gregg, R., Righetti, L., Buchli, J., Schaal, S.

In 2010 10th IEEE-RAS International Conference on Humanoid Robots, pages: 1-7, IEEE, Nashville, USA, 2010 (inproceedings)

Abstract
Energy-shaping control methods have produced strong theoretical results for asymptotically stable 3D bipedal dynamic walking in the literature. In particular, geometric controlled reduction exploits robot symmetries to control momentum conservation laws that decouple the sagittal-plane dynamics, which are easier to stabilize. However, the associated control laws require high-dimensional matrix inverses multiplied with complicated energy-shaping terms, often making these control theories difficult to apply to highly-redundant humanoid robots. This paper presents a first step towards the application of energy-shaping methods on real robots by casting controlled reduction into a framework of constrained accelerations for inverse dynamics control. By representing momentum conservation laws as constraints in acceleration space, we construct a general expression for desired joint accelerations that render the constraint surface invariant. By appropriately choosing an orthogonal projection, we show that the unconstrained (reduced) dynamics are decoupled from the constrained dynamics. Any acceleration-based controller can then be used to stabilize this planar subsystem, including passivity-based methods. The resulting control law is surprisingly simple and represents a practical way to employ control theoretic stability results in robotic platforms. Simulated walking of a 3D compass-gait biped show correspondence between the new and original controllers, and simulated motions of a 16-DOF humanoid demonstrate the applicability of this method.

am mg

link (url) DOI [BibTex]

link (url) DOI [BibTex]


no image
Surface tension driven water strider robot using circular footpads

Ozcan, O., Wang, H., Taylor, J. D., Sitti, M.

In Robotics and Automation (ICRA), 2010 IEEE International Conference on, pages: 3799-3804, 2010 (inproceedings)

pi

[BibTex]

[BibTex]


no image
Atomic-Force-Microscopy-Based Nanomanipulation Systems

Onal, C. D., Ozcan, O., Sitti, M.

In Handbook of Nanophysics: Nanomedicine and Nanorobotics, pages: 1-15, CRC Press, 2010 (incollection)

pi

[BibTex]

[BibTex]


no image
Variable impedance control - a reinforcement learning approach

Buchli, J., Theodorou, E., Stulp, F., Schaal, S.

In Robotics Science and Systems (2010), Zaragoza, Spain, June 27-30, 2010, clmc (inproceedings)

Abstract
One of the hallmarks of the performance, versatility, and robustness of biological motor control is the ability to adapt the impedance of the overall biomechanical system to different task requirements and stochastic disturbances. A transfer of this principle to robotics is desirable, for instance to enable robots to work robustly and safely in everyday human environments. It is, however, not trivial to derive variable impedance controllers for practical high DOF robotic tasks. In this contribution, we accomplish such gain scheduling with a reinforcement learning approach algorithm, PI2 (Policy Improvement with Path Integrals). PI2 is a model-free, sampling based learning method derived from first principles of optimal control. The PI2 algorithm requires no tuning of algorithmic parameters besides the exploration noise. The designer can thus fully focus on cost function design to specify the task. From the viewpoint of robotics, a particular useful property of PI2 is that it can scale to problems of many DOFs, so that RL on real robotic systems becomes feasible. We sketch the PI2 algorithm and its theoretical properties, and how it is applied to gain scheduling. We evaluate our approach by presenting results on two different simulated robotic systems, a 3-DOF Phantom Premium Robot and a 6-DOF Kuka Lightweight Robot. We investigate tasks where the optimal strategy requires both tuning of the impedance of the end-effector, and tuning of a reference trajectory. The results show that we can use path integral based RL not only for planning but also to derive variable gain feedback controllers in realistic scenarios. Thus, the power of variable impedance control is made available to a wide variety of robotic systems and practical applications.

am

link (url) [BibTex]

link (url) [BibTex]


no image
Inverse dynamics with optimal distribution of ground reaction forces for legged robot

Righetti, L., Buchli, J., Mistry, M., Schaal, S.

In Proceedings of the 13th International Conference on Climbing and Walking Robots (CLAWAR), pages: 580-587, Nagoya, Japan, sep 2010 (inproceedings)

Abstract
Contact interaction with the environment is crucial in the design of locomotion controllers for legged robots, to prevent slipping for example. Therefore, it is of great importance to be able to control the effects of the robots movements on the contact reaction forces. In this contribution, we extend a recent inverse dynamics algorithm for floating base robots to optimize the distribution of contact forces while achieving precise trajectory tracking. The resulting controller is algorithmically simple as compared to other approaches. Numerical simulations show that this result significantly increases the range of possible movements of a humanoid robot as compared to the previous inverse dynamics algorithm. We also present a simplification of the result where no inversion of the inertia matrix is needed which is particularly relevant for practical use on a real robot. Such an algorithm becomes interesting for agile locomotion of robots on difficult terrains where the contacts with the environment are critical, such as walking over rough or slippery terrain.

am mg

DOI [BibTex]

DOI [BibTex]


no image
Effects of Wheel Synchronization for the Hybrid Leg-Wheel Robot Asguard

Babu, A., Joyeux, S., Schwendner, J., Grimminger, F.

In Proceedings of International Symposium on Artificial Intelligence, Robotics and Automation in Space. International Symposium on Artificial Intelligence, Robotics and Automation in Space (iSAIRAS-10), August 29-September 1, Sapporo, Japan, August 2010 (inproceedings)

am

[BibTex]

[BibTex]

2008


no image
Simulation and analysis of a passive pitch reversal flapping wing mechanism for an aerial robotic platform

Arabagi, V., Sitti, M.

In Intelligent Robots and Systems, 2008. IROS 2008. IEEE/RSJ International Conference on, pages: 1260-1265, 2008 (inproceedings)

pi

Project Page [BibTex]

2008


Project Page [BibTex]


no image
Human movement generation based on convergent flow fields: A computational model and a behavioral experiment

Hoffmann, H., Schaal, S.

In Advances in Computational Motor Control VII, Symposium at the Society for Neuroscience Meeting, Washington DC, 2008, 2008, clmc (inproceedings)

am

link (url) [BibTex]

link (url) [BibTex]


no image
Fabrication and Characterization of Biologically Inspired Mushroom-Shaped Elastomer Microfiber Arrays

Kim, S., Sitti, M.

In ASME 2008 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, pages: 839-847, 2008 (inproceedings)

pi

Project Page [BibTex]

Project Page [BibTex]


no image
Gecko inspired micro-fibrillar adhesives for wall climbing robots on micro/nanoscale rough surfaces

Aksak, B., Murphy, M. P., Sitti, M.

In Robotics and Automation, 2008. ICRA 2008. IEEE International Conference on, pages: 3058-3063, 2008 (inproceedings)

pi

Project Page [BibTex]

Project Page [BibTex]