IROS 2015 论文列表

Visual recognition in multi-robot systems is afflicted with a peculiar problem that observations made from different viewpoints bring different perspectives. Due to the lack of a discriminable representation between a target and its surroundings in different viewpoints, realizing cognitive sharing of the object among the robots in an unconstructed environment is a challenging issue. In this paper, we propose novel description algorithms of the target representation based on ambiguity minimization of peripheral context. The target is represented by a labeled-graph and its structure is determined by minimizing a metric of representational ambiguity using an entropy evaluation with metaheuristics. Experimental results show the significantly improvement of cognitive sharing by the proposed method.

In co-manipulation, humans and robots solve manipulation tasks together. Virtual guides are important tools for co-manipulation, as they constrain the movement of the robot to avoid undesirable effects, such as collisions with the environment. Defining virtual guides is often a laborious task requiring expert knowledge. This restricts the usefulness of virtual guides in environments where new tasks may need to be solved, or where multiple tasks need to be solved sequentially, but in an unknown order. To this end, we propose a framework for multiple probabilistic virtual guides, and demonstrate a concrete implementation of such guides using kinesthetic teaching and Gaussian mixture models. Our approach enables non-expert users to design virtual guides through demonstration. Also, they may demonstrate novel guides, even if already known guides are active. Finally, users are able to intuitively select the appropriate guide from a set of guides through physical interaction with the robot. We evaluate our approach in a pick-and-place task, where users are to place objects at one of several positions in a cupboard.

Autonomous vehicles require advances sensing technologies, in order to be able to safely share the environment with human operators. Those sensing technologies are in fact necessary for identifying the presence of unforeseen objects, and measuring their position and velocity. Furthermore, classification is necessary for effectively predicting their behavior. In this paper we consider the presence of sensing systems both on-board each vehicle, and installed on infrastructural elements. While the simultaneous presence of multiple sources of information heavily improves the amount (and quality) of available data, it generates the need for effective data fusion and storage systems. Hence, we introduce a centralized cloud service, that is in charge of receiving and merging data acquired by different sensing systems. Those data are then distributed to the autonomous vehicles, that exploit them for implementing advanced navigation strategies. The proposed methodology is validated in a real industrial environment to safely perform obstacle avoidance with an autonomously driven forklift.

Real-time estimation of human-cloth relationship is crucial for efficient learning of motor skills in robotic clothing assistance. However, cloth state estimation using a depth sensor is a challenging problem with inherent ambiguity. To address this problem, we propose the offline learning of a cloth dynamics model by incorporating reliable motion capture data and applying this model for the online tracking of human-cloth relationship using a depth sensor. In this study, we evaluate the performance of using a shared Gaussian Process Latent Variable Model in learning the dynamics of clothing articles. The experimental results demonstrate the effectiveness of shared GP-LVM in capturing cloth dynamics using few data samples and the ability to generalize to unseen settings. We further demonstrate three key factors that affect the predictive performance of the trained dynamics model.

Navigating non-holonomic mobile robots in dynamic environments is challenging because it requires computing at each instant, the space of collision free velocities, characterized by a set of highly non-linear and non-convex inequalities. Moreover, uncertainty in obstacle trajectories further increases the complexity of the problem, as it now becomes imperative to relate the space of collision free velocities to a confidence measure. In this paper, we present a novel perspective towards analyzing and solving probabilistic collision avoidance constraints based on our previous works on non-linear time scaling. In particular, we have shown earlier that a time scaled version of collision cone constraints can be solved in closed form and thus can be used to efficiently characterize the space of collision free velocities. In the current proposed work, we present a probabilistic version of time scaled collision cone constraints obtained by representing obstacle states through generic probability distributions. We present a novel reformulation of the probabilistic constraints into a family of deterministic algebraic constraints. The solution space of each member of the family can be derived in closed form and at the same time, can also be related to the lower bound on confidence measure through Cantelli's inequality. Thus, the proposed work represents a significant improvement over the current state of the art frameworks where probabilistic collision avoidance constraints are solved through exhaustive sampling in the state-control space. We also present a cost metric which serves as the basis for the construction of the various collision avoidance maneuvers based on factors like deviation from the current path, acceleration/de-acceleration capability of the robot, confidence of collision avoidance etc. We very briefly explain how the current robot state can be connected to the solution space of safe velocities in smooth time optimal fashion. Finally, the validity of the prop... Show More

Many household chores and industrial manufacturing tasks require a certain compliant behavior to make deliberate physical contact with the environment. This compliant behavior can be implemented by modern robotic manipulators. However, in order to plan the task execution, a robot requires generic process models of these tasks which can be adapted to different domains and varying environmental conditions. In this work we propose a classification of compliant manipulation tasks meeting these requirements, to derive related actions for automated planning. We also present a classification for the sub-category of wiping tasks, which are most common and of great importance in service robotics.We categorize actions from an object-centric perspective to make them independent of any specific robot kinematics. The aim of the proposed taxonomy is to guide robotic programmers to develop generic actions for any kind of robotic systems in arbitrary domains.

Determining and detecting risk markers for mental illness remains a labor intensive process, requiring vast amounts of observations by clinical professionals. Motor stereotypies, which are defined as involuntary repetitive motor behaviors, invariant in form, that, to an observer, appear to serve no purpose, are a class of risk markers which are very amenable to video analysis. These behaviors are associated with mental illnesses such as Autism, Rett Syndrome, and other developmental disabilities. This paper investigates the application of innovative automated methods to recognize these subtle motor indicators. To train and test our methods, a dataset of actions resembling motor stereotypies was created by engaging the normally developing children at the University of Minnesota's Shirley G. Moore Laboratory School. Comparison to a publicly available dataset depicting a subset of behaviors is performed as well. This work demonstrates the applicability of various techniques in the behavioral science domain. The results show that these techniques can perform well on a difficult and challenging real-world scenario.

To enable robots to learn handover orientations from observing natural handovers, we conduct a user study to measure and compare natural handover orientations with giver-centered and receiver-centered handover orientations for twenty common objects. We use a distance minimization approach to compute mean handover orientations. We posit that, computed means of receiver-centered orientations could be used by robot givers to achieve more efficient and socially acceptable handovers. Furthermore, we introduce the notion of affordance axes for comparing handover orientations, and offer a definition for computing them. Observable patterns were found in receiver-centered handover orientations. Comparisons show that depending on the object, natural handover orientations may not be receiver-centered; thus, robots may need to distinguish between good and bad handover orientations when learning from natural handovers.

To realize the underwater intervention, a Father-son Underwater Intervention Robotic System (FUIRS) is proposed in our laboratory. The FUIRS employs a novel biomimetic microrobot to realize underwater manipulation tasks. The main work of this paper is to describe the biomimetic microrobot which is inspired by an octopus. The microrobot is mainly composed of body and arm. We use IPMC to fabricate the arms. To investigate the most suitable arm shape, three commonly used arms were designed, i.e., regular rectangle shape, human finger shape and tapered shape. After analyzing the force and deflection features, the human finger shape arm was selected to be installed on the microrobot. The robot body is fabricated by the 3D printer. A micro proximity sensor was installed on it as the eyes. To enhance the payload of the microrobot, a novel buoyancy adjustment method was proposed which can provide 11.8mN additional buoyancy force. Finally, a series of underwater manipulation experiments were carried out and the results indicated that the biomimetic microrobot can realize underwater manipulation for the targets with different shape and size objects successfully.

This paper proposes a novel Cosserat rod model based method to estimate contact forces based on the shape analysis of the catheter. We simplify the original Cosserat rod model to achieve real-time estimation of the force magnitude and directions applied to the catheter tip. The simplified model contains a set of arithmetic equations, which can be rapidly solved using an iterative optimization algorithm. Experimental evaluation shows that the computational frequency of the force estimation was found to be 33.7Hz. Both the magnitude and the direction of the contact force were accurately estimated. The accuracy of the estimations of magnitude was 89.50%, and for the contact direction was 88.13%. Mean errors of the contact force and the contact angle are 7 × 10-4N and 0.931 degree respectively. To prove the concept, the catheter shape is detected through a RGB camera. However, the proposed method can be easily applied to existing catheter gating and detecting methods using the medical imaging environments such as X-ray fluoroscopy, CT, ultrasound, magnetic field and MRI. The results of evaluation experiments demonstrate what the proposed method is promising for force estimation without the need of a physical force sensor in various types of catheter tips.

Carrying unknown heavy payload is one of the classical problems in robotics. To realize heavy payload capability, the use of high reduction gears, which causes large motor-side inertia and large friction, is inevitable. This paper proposes a disturbance observer (DOB)-based approach that (1) reduces highly amplified motor-side inertia (2) compensates for large friction (3) compensates for the unknown heavy payload without sensory information. However, in principle, the DOB is applicable only for linear system. To overcome this, a concept of effective joint torque is introduced, which allows us to separate linear motor-side dynamics from the nonlinear robot dynamics. As a result, it becomes possible to apply high order DOB to compensate for the uncertainties. Mathematical analysis shows closed-loop stability, optimality, and passivity. Experiments and simulations are performed to verify the method.

Pheromone-based communication is one of the most effective ways of communication widely observed in nature. It is particularly used by social insects such as bees, ants and termites; both for inter-agent and agent-swarm communications. Due to its effectiveness; artificial pheromones have been adopted in multi-robot and swarm robotic systems for more than a decade. Although, pheromone-based communication was implemented by different means like chemical (use of particular chemical compounds) or physical (RFID tags, light, sound) ways, none of them were able to replicate all the aspects of pheromones as seen in nature. In this paper, we propose a novel artificial pheromone system that is reliable, accurate and it uses off-the-shelf components only - LCD screen and low-cost USB camera. The system allows to simulate several pheromones and their interactions and to change parameters of the pheromones (diffusion, evaporation, etc.) on the fly allowing for controllable experiments. We tested the performance of the system using the Colias platform in single-robot and swarm scenarios. To allow the swarm robotics community to use the system for their research, we provide it as a freely available open-source package.

An object recognition engine needs to extract discriminative features from data representing an object and accurately classify the object to be of practical use in robotics. Furthermore, the classification of the object must be rapidly performed in the presence of a voluminous stream of data. These conditions call for a distributed and scalable architecture that can utilize a cloud computing infrastructure for performing object recognition. This paper introduces a Cloud-based Object Recognition Engine (CORE) to address these needs. CORE is able to train on large-scale datasets, perform classification of 3D point cloud data, and efficiently transfer data in a robotic network.

A novel Kalman filter to estimate the center of mass (COM) of a Humanoid robot is proposed. In the conventional works, COM was estimated by some methods. First one is the kinematics computation based on the mass property of the robot and the global position and attitude of the body frame, but those errors degrade the estimation accuracy. Second is the double integral of COM acceleration computed by the measured external force. However, its accuracy suffers from the error accumulation with the integration and the initial error remains. Third is based on the relationship between COM and the zero-moment point (ZMP), but it ignores the torque around COM. Additionally, it only dealt with the horizontal movement. For those problems, the proposed method combines those informations in order to improve the accuracy. Particularly, in order to estimate three dimensional motion of COM, the proposed method reduces the offset included in the vertical component by utilizing the interference between the horizontal and vertical component of COM shown in third information. Through the simulation, the improvement by the proposed method is verified.

This paper presents the autonomous tracking and following of a marine vessel by an Unmanned Surface Vehicle in the presence of dynamic obstacles while following the International Regulations for Preventing Collisions at Sea (COLREGS) rules. The motion prediction for the target vessel is based on Monte-Carlo sampling of dynamically feasible and collision-free paths with fuzzy weights, leading to a predicted path resembling anthropomorphic driving behavior. This prediction is continuously optimized for a particular target by learning the necessary parameters for a 3-degree-of-freedom model of the vessel and its maneuvering behavior from its path history without any prior knowledge. The path planning for the USV with COLREGS is achieved on a grid-based map in a single stage by incorporating A* path planning with Artificial Terrain Costs for dynamically changing obstacles. Various scenarios for interaction, including multiple civilian and adversarial vessels, are handled by the planner with ease. The effectiveness of the algorithms has been demonstrated both in representative simulations and on-water experiments.

In this paper, we propose a novel approach to generate a temporal consistent semantic map for 3D indoor scenes. In contrast to previous techniques which generate the semantic map on the whole global scene immediately or smooth the semantic predictions by a graph model, we intend to discover temporal information over RGB-D images and leverage it to enforce the label consistency. Our method contains two key components: a low-level component and a high-level component. On the low level, temporal segmentation method is adopted to find the correspondence of the superpixels incrementally. On the high level, temporal information is treated as the higher order cliques and a higher-order Dense Conditional Random Fields (CRF) is utilized to jointly infer the object category and structural class of the global point cloud. On the experiments, we compare our temporal consistent segmentation algorithm with the state-of-the-art approach and generate the semantic maps from the NYU v2 dataset. Our experiments demonstrate that temporal consistent constraints are significant for the semantic mapping procedure and can improve the precision of the semantic mapping results.

This paper is concerned with the interpretation of visual information for robot localization. It presents a probabilistic localization system that generates an appropriate observation model online, unlike existing systems which require pre-determined belief models. This paper proposes that probabilistic visual localization requires two major operating modes - one to match locations under similar conditions and the other to match locations under different conditions. We develop dual observation likelihood models to suit these two different states, along with a similarity measure-based method that identifies the current conditions and switches between the models. The system is experimentally tested against different types of ongoing appearance change. The results demonstrate that the system is compatible with a wide range of visual front-ends, and the dual-model system outperforms a single-model or pre-trained approach and state-of-the-art localization techniques.

Navigating in a previously unknown environment and recognizing naturally occurring text in a scene are two important autonomous capabilities that are typically treated as distinct. However, these two tasks are potentially complementary, (i) scene and pose priors can benefit text spotting, and (ii) the ability to identify and associate text features can benefit navigation accuracy through loop closures. Previous approaches to autonomous text spotting typically require significant training data and are too slow for real-time implementation. In this work, we propose a novel high-level feature descriptor, the “junction”, which is particularly well-suited to text representation and is also fast to compute. We show that we are able to improve SLAM through text spotting on datasets collected with a Google Tango, illustrating how location priors enable improved loop closure with text features.

This paper introduces a Biologically Inspired Dead-beat (BID) controller for bipedal running in 3D. The controller runs in real-time, is extremely robust against perturbations and allows for versatile running patterns. It is based on the encoding of leg forces and CoM trajectories during stance as polynomial splines, allowing for intuitive and primarily analytical controller design. The performance of the control framework is tested in various simulations for a bipedal point-mass model.