IROS 2015 论文列表

One of the major goals for mobile robots is to be able to traverse any kind of terrains. A possible way to achieve this goal is by the use of legged robots, as they have increased mobility. However, this would require them to be able to modify their gaits, based on the identification of the terrain that they are currently traversing. In this paper, we introduce a number of novel methods to address this issue of autonomous terrain classification and clustering, based on tactile data collected with a walking robot. The proposed learning methods are based on the Pitman-Yor process mixture of Gaussians, a Bayesian nonparametric prior, well-suited for density estimation. This model is initially used to learn the non-Gaussian distribution of the features produced from proprioceptive (force/torque) signals from the legs, registered during the interaction of one robot foot with a terrain. Then, we exploit its capacity on clustering and discovering structures in the data to identify terrains in the feature space. Experiments were conducted on a six-legged robot, thus demonstrating the applicability of the Pitman-Yor process mixture of Gaussians for terrain identification. In particular, we obtained a classification success rate of 82% and 51% accuracy, with our supervised learning and unsupervised learning approach respectively.

We propose a method for generating interpretable descriptions of inputs that cause faults in high-dimensional software interfaces. Our method models the set of fault-triggering inputs as a Cartesian product and identifies this set by actively querying the system under test. The active sampling scheme is very efficient in the common case that few fields in the interface are relevant to causing the fault. This scheme also solves the problem of efficiently finding sufficient examples to model rare faults, which is problematic for other learning-based methods. Compared to other techniques, ours requires no parameter turning or post-processing in order to produce useful results. We analyze the method qualitatively, theoretically, and empirically. An experimental evaluation demonstrates superior performance and reliability compared to a basic decision tree approach. We also briefly discuss how the method has assisted in debugging a commercial autonomous ground vehicle system.

With robots leaving factory environments and entering less controlled domains, possibly sharing living space with humans, safety needs to be guaranteed. To this end, some form of awareness of their body surface and the space surrounding it is desirable. In this work, we present a unique method that lets a robot learn a distributed representation of space around its body (or peripersonal space) by exploiting a whole-body artificial skin and through physical contact with the environment. Every taxel (tactile element) has a visual receptive field anchored to it. Starting from an initially blank state, the distance of every object entering this receptive field is visually perceived and recorded, together with information whether the object has eventually contacted the particular skin area or not. This gives rise to a set of probabilities that are updated incrementally and that carry information about the likelihood of particular events in the environment contacting a particular set of taxels. The learned representation naturally serves the purpose of predicting contacts with the whole body of the robot, which is of clear behavioral relevance. Furthermore, we devised a simple avoidance controller that is triggered by this representation, thus endowing a robot with a “margin of safety” around its body. Finally, simply reversing the sign in the controller we used gives rise to simple “reaching” for objects in the robot's vicinity, which automatically proceeds with the most activated (closest) body part.

In this paper, an approach for learning optimal striking points is proposed. Based on a ball-flight model and a rebound model, a set of reachable striking points within the robot's workspace can be obtained. However, while these striking points are geometrically reachable, their success probability differs substantially due to the robot's nonlinear dynamics, the distance to the ball, the need to reach sufficient velocity as well as the right angle at interception and non-uniform sensitivity to errors. Thus, it is crucial for a ping-pong robotic system to select striking points well. As a successful ball interception is the result of various factors that cannot be modeled straightforwardly, we suggest determining optimal striking points based on a reward function that measures how well the ping-pong ball's trajectory and the racket's movement coincidence. In this approach, we propose to learn a stochastic policy over the reward given the prospective striking point in order to facilitate exploration of a wide range of prospective striking points. The resulting learning method takes both the amount of experience data and its confidence into account to reach optimal solutions reliably. Evaluation with a real robotic system demonstrates the applicability of the proposed method.

Human-robot collaboration seeks to have humans and robots closely interacting in everyday situations. For some tasks, physical contact between the user and the robot may occur, originating significant challenges at safety, cognition, perception and control levels, among others. This paper focuses on robot motion adaptation to parameters of a collaborative task, extraction of the desired robot behavior, and variable impedance control for human-safe interaction. We propose to teach a robot cooperative behaviors from demonstrations, which are probabilistically encoded by a task-parametrized formulation of a Gaussian mixture model. Such encoding is later used for specifying both the desired state of the robot, and an optimal feedback control law that exploits the variability in position, velocity and force spaces observed during the demonstrations. The whole framework allows the robot to modify its movements as a function of parameters of the task, while showing different impedance behaviors. Tests were successfully carried out in a scenario where a 7 DOF backdrivable manipulator learns to cooperate with a human to transport an object.

Learning models of behaviours has many applications in robotics spanning both control, e.g. learning from demonstration and perception, e.g. monitoring and surveillance. Inverse reinforcement learning encodes behaviours as a reward function learned from a set of demonstrations. This paper addresses the problem of learning from unlabelled datasets containing an unknown number of behaviours in continuous action-state spaces. The proposed method uses a hierarchical clustering approach to directly group trajectories that share a common reward function. The similarity metric is based on the distribution of maximum entropy of the feature counts computed using path integrals. We evaluated the method in three different tasks: navigation on a set of synthetic maps, human driving styles on a simulator and human reaching. Results show that clustering in the reward space is able to discover the latent reward structure resulting in compact models that can generate all the observed behaviours.

Learning motor skills from multiple demonstrations presents a number of challenges. One of those challenges is the occurrence of occlusions and lack of sensor coverage, which may corrupt part of the recorded data. Another issue is the variability in speed of execution of the demonstrations, which may require a way of finding the correspondence between the time steps of the different demonstrations. In this paper, an approach to learn motor skills is proposed that accounts both for spatial and temporal variability of movements. This approach, based on an Expectation-Maximization algorithm to learn Probabilistic Movement Primitives, also allows for learning motor skills from partially observed demonstrations, which may result from occlusion or lack of sensor coverage. An application of the algorithm proposed in this work lies in the field of Human-Robot Interaction when the robot has to react to human movements executed at different speeds. Experiments in which a robotic arm receives a cup handed over by a human illustrate this application. The capabilities of the algorithm in learning and predicting movements are also evaluated in experiments using a data set of letters and a data set of golf putting movements.

Although the neurological impairments of Parkinson's disease (PD) patients are well known to go along with motor control deficits, e.g., tremor, rigidity, and reduced movement, not much is known about the motor control parameters affected by the disease. In this paper, we therefore present a novel approach to human motions analysis using motor control strategies with joint weight parameterization. We record the motions of healthy subjects and PD patients performing a hand coordination task with the whole-body XSens MVN motion capture system. For our motion strategy analysis we then follow a two step approach. First, we perform a complexity reduction by mapping the recorded human motions to a simplified kinematic model of the upper body. Second, we reproduce the recorded motions using a Jacobian weighted damped least squares controller with adaptive joint weights. We developed a method to iteratively learn the joint weights of the controller with the mapped human joint trajectories as reference input. Finally, we use the learned joint weights for a quantitative comparison between the motion control strategies of healthy subjects and PD patients. Other than expected from clinical experience, we found that the joint weights are almost evenly distributed along the arm in the PD group. In contrast to that, the proximal joint weights of the healthy subjects are notably larger than the distal ones.

The number of mechanical degrees of freedom (DoFs) within rehabilitation robots directly influences the scope of the movements that a subject can perform when training walking. Currently, gait rehabilitation robots have a limited number of mechanical DoFs, as a consequence this limits the movements these robots can make possible. In this paper, the novel gait rehabilitation system CORBYS is presented which consists of the mobile platform and a powered orthosis which is attached to the platform. The CORBYS powered orthosis has 16 DoFs enabling more physiological movements, making it a state-of-the-art gait rehabilitation robotic system. With the sufficient number of DoFs to enable natural-like walking, the CORBYS robotic system enables the integration of the “learning gait by therapist demonstration” paradigm. This paper presents the fully integrated functional CORBYS gait rehabilitation system, with the focus on the implementation aspects which enable generation of the reference gait trajectory through learning by therapist demonstration, and the use of the generated trajectory in the robotic therapy session. The results of the initial evaluation of the robotic system obtained in tests with a selected patient are given in the paper.

Learning from demonstration by means of non-rigid point cloud registration is an effective tool for learning to manipulate a wide range of deformable objects. However, most methods that use non-rigid registration to transfer demonstrated trajectories assume that the test and demonstration scene are structurally very similar, with any variation explained by a non-linear transformation. In real-world tasks with clutter and distractor objects, this assumption is unrealistic. In this work, we show that a trajectory-aware non-rigid registration method that uses multiple demonstrations to focus the registration process on points that are relevant to the task can effectively handle significantly greater visual variation than prior methods that are not trajectory-aware. We demonstrate that this approach achieves superior generalization on several challenging tasks, including towel folding and grasping objects in a box containing irrelevant distractors.

This paper describes a vision-based method of guiding autonomous vehicles within crop rows in agricultural fields where the crop rows are challenging to detect or their appearance is not known a-priori. The location of the crop rows is estimated with an SVM regression algorithm using colour, texture and 3D structure descriptors from a forward facing stereo camera pair. Our system rapidly learns a model online with minimal user input, and then uses this model to track crop rows. Results demonstrate our method is able to learn and track a wide variety of crops with an RMS error of less than 3cm. We also present online control results demonstrating our system autonomously steering a robot for 3km.

Applications of reinforcement learning for robotic manipulation often assume an episodic setting. However, controllers trained with reinforcement learning are often situated in the context of a more complex compound task, where multiple controllers might be invoked in sequence to accomplish a higher-level goal. Furthermore, training such controllers typically requires resetting the environment between episodes, which is typically handled manually. We describe an approach for training chains of controllers with reinforcement learning. This requires taking into account the state distributions induced by preceding controllers in the chain, as well as automatically training reset controllers that can reset the task between episodes. The initial state of each controller is determined by the controller that precedes it, resulting in a non-stationary learning problem. We demonstrate that a recently developed method that optimizes linear-Gaussian controllers under learned local linear models can tackle this sort of non-stationary problem, and that training controllers concurrently with a corresponding reset controller only minimally increases training time. We also demonstrate this method on a complex tool use task that consists of seven stages and requires using a toy wrench to screw in a bolt. This compound task requires grasping and handling complex contact dynamics. After training, the controllers can execute the entire task quickly and efficiently. Finally, we show that this method can be combined with guided policy search to automatically train nonlinear neural network controllers for a grasping task with considerable variation in target position.

Very often, when addressing the problem of human-robot skill transfer in task space, only the Cartesian position of the end-effector is encoded by the learning algorithms, instead of the full pose. However, orientation is just as important as position, if not more, when it comes to successfully performing a manipulation task. In this paper, we present a framework that allows robots to learn the full poses of their end-effectors in a task-parameterized manner. Our approach permits the encoding of complex skills, such as those found in bimanual manipulation scenarios, where the generalized coordination patterns between end-effectors (i.e. position and orientation patterns) need to be considered. The proposed framework combines a dynamical systems formulation of the demonstrated trajectories, both in ℝ3 and SO(3), and task-parameterized probabilistic models that build local task representations in both spaces, based on which it is possible to extract the relevant features of the demonstrated skill. We validate our approach with an experiment in which two 7-DoF WAM robots learn to perform a bimanual sweeping task.

Predicting the outcome of an action can help a robot detect failures in advance, and schedule action replanning before an error occurs. We propose using an interactive physics based simulator with the aim of collecting realistic data to be used for learning. We then show how we save and query for specific information from the data more effectively. The data from the simulation is used to learn a failure detection model which is utilized by a real robot performing the same actions. We show that learning from simulation data is realistic enough to be applied on a real robot. The learning algorithm is more simple in design and outperforms the more complex one from our previous work.

Predictive State Representations (PSRs) allow modeling of dynamical systems directly in observables and without relying on latent variable representations. A problem that arises from learning PSRs is that it is often hard to attribute semantic meaning to the learned representation. This makes generalization and planning in PSRs challenging. In this paper, we extend PSRs and introduce the notion of PSRs that include prior information (P-PSRs) to learn representations which are suitable for planning and interpretation. By learning a low-dimensional embedding of test features we map belief points of similar semantic to the same region of a subspace. This facilitates better generalization for planning and semantical interpretation of the learned representation. In specific, we show how to overcome the training sample bias and introduce feature selection such that the resulting representation emphasizes observables related to the planning task. We show that our P-PSRs result in qualitatively meaningful representations and present quantitative results that indicate improved suitability for planning.

A key challenge of developing robots that work closely with people is creating a user interface that allows a user to communicate complex instructions to a robot quickly and easily. We consider a walking logistics support robot, which is designed to carry heavy loads to locations that are too difficult to reach with a wheeled or tracked vehicle. In this application the robot is carrying equipment and supplies for a group of pedestrians, and the primary task for the user interface is to keep the robot traveling with the overall group in the right formation. This paper presents a marker tracking system that uses near infrared cameras, retro-reflective markers, and LIDAR to allow a particular user to designate himself as the robot's leader, and guide the robot along a desired path. We provide an extensive quantitative evaluation to show that the proposed system is able to detect and track a leader through unconstrained and cluttered off-road environments under a wide variety of illumination and motion conditions.

Nowadays real time visual Simultaneous Localization And Mapping (SLAM) algorithms exist and rely on consistent measurements across multiple views. In indoor environments, where majority of robot's activity takes place, severe occlusions can occur, e.g., when turning around a corner or moving from one room to another. In these situations, SLAM algorithms can not establish correspondences across views, which leads to failures in camera localization or map construction.

We propose a novel Large-Scale Direct SLAM algorithm for stereo cameras (Stereo LSD-SLAM) that runs in real-time at high frame rate on standard CPUs. In contrast to sparse interest-point based methods, our approach aligns images directly based on the photoconsistency of all high-contrast pixels, including corners, edges and high texture areas. It concurrently estimates the depth at these pixels from two types of stereo cues: Static stereo through the fixed-baseline stereo camera setup as well as temporal multi-view stereo exploiting the camera motion. By incorporating both disparity sources, our algorithm can even estimate depth of pixels that are under-constrained when only using fixed-baseline stereo. Using a fixed baseline, on the other hand, avoids scale-drift that typically occurs in pure monocular SLAM.We furthermore propose a robust approach to enforce illumination invariance, capable of handling aggressive brightness changes between frames - greatly improving the performance in realistic settings. In experiments, we demonstrate state-of-the-art results on stereo SLAM benchmarks such as Kitti or challenging datasets from the EuRoC Challenge 3 for micro aerial vehicles.

We propose a real-time, direct monocular SLAM method for omnidirectional or wide field-of-view fisheye cameras. Both tracking (direct image alignment) and mapping (pixel-wise distance filtering) are directly formulated for the unified omnidirectional model, which can model central imaging devices with a field of view above 180 °. This is in contrast to existing direct mono-SLAM approaches like DTAM or LSD-SLAM, which operate on rectified images, in practice limiting the field of view to around 130 ° diagonally. Not only does this allows to observe - and reconstruct - a larger portion of the surrounding environment, but it also makes the system more robust to degenerate (rotation-only) movement. The two main contribution are (1) the formulation of direct image alignment for the unified omnidirectional model, and (2) a fast yet accurate approach to incremental stereo directly on distorted images. We evaluated our framework on real-world sequences taken with a 185 ° fisheye lens, and compare it to a rectified and a piecewise rectified approach.