Contact-State Segmentation Using Particle Filters for Programming by Human Demonstration in Compliant-Motion Tasks

This paper presents a contribution to programming by human demonstration, in the context of compliant-motion task specification for sensor-controlled robot systems that physically interact with the environment. One wants to learn about the geometric parameters of the task and segment the total motion executed by the human into subtasks for the robot, that can each be executed with simple compliant-motion task specifications. The motion of the human demonstration tool is sensed with a 3-D camera, and the interaction with the environment is sensed with a force sensor in the human demonstration tool. Both measurements are uncertain, and do not give direct information about the geometric parameters of the contacting surfaces, or about the contact formations (CFs) encountered during the human demonstration. The paper uses a Bayesian sequential Monte Carlo method (also known as a particle filter) to do the simultaneous estimation of the CF (discrete information) and the geometric parameters (continuous information). The simultaneous CF segmentation and the geometric parameter estimation are helped by the availability of a contact state graph of all possible CFs. The presented approach applies to all compliant-motion tasks involving polyhedral objects with a known geometry, where the uncertain geometric parameters are the poses of the objects. This work improves the state of the art by scaling the contact estimation to all possible contacts, by presenting a prediction step based on the topological information of a contact state graph, and by presenting efficient algorithms that allow the estimation to operate in real time. In real-world experiments, it is shown that the approach is able to discriminate in real time between some 250 different CFs in the graph     

3-D mapping of natural environments with trees by means of mobile perception

In this paper, a method of generating a three-dimensional (3-D) geometric model for large-scale natural environments with trees is presented. The environment mapping method, which uses range images as measurement data, consists of three main phases. First, geometric feature objects are extracted from each of the range images. Second, the relative coordinate transformations (i.e., registrations) between the sensor viewpoint locations, where the range data are measured, are computed. Third, an integrated map is formed by transforming the submap data into a common frame of reference. Tree trunks visible in the range images are modeled with cylinder segments and utilized as reference features for registration computation. The final integrated 3-D model consists of the cylinder segments representing the visible sections of the tree trunks, as well as of the ground elevation data. The constructed environment map can be utilized as, for example, a virtual task environment for outdoor robotic machines such as new-generation forest working machines or service robots.     

Reasoning about nature of contact for planning manipulators tasks

Extracting information about contact between two convex bodies from the measured force vector is a prerequisite for any fine compliant motion control strategy. Contact information contains the direction and orientation of the contact surface normal and its relative location and orientation with respect to the compliant reference frame system.A method for interpreting the contact force feedback during compliant robot motion control, using kinematic screws, is presented. Domain specific rules combined with partial a priori knowledge of mating parts geometry and interpreted force signals are used to reason and make inferences about the initial contact configuration. The likely contact surfaces are predicted and point(s) or line(s) of contact are fully defined. These surfaces are idealized and represented by quadratic equations or polyhedral surfaces. The geometric properties of surfaces at the contact location are used to select the contact configuration when multiple solutions exist.An algorithm for predicting the Expected Contact Configuration (ECC) has been developed and is illustrated here with examples. Experimental validation of the developed expert system prototype, using a 6R manipulator, a six-axis force sensor, and a host computer is described.     

Evaluating part designs for automatic feeding in robotic assembly systems: A geometric analysis of the nesting of polyhedral objects

One of the problems encountered by automatic feeding devices, such as vibratory bowl feeders, in robotic assembly systems is nesting which leads to inseparability of parts and jamming. In this paper, we describe a design evaluation procedure to determine whether a part is prone to nesting and to quantify the degree of nesting. The proposed methodology reduces the 3-D protrusion-hole polyhedral containment problem of nesting to a 2-D polygon containment problem by employing a loop-based feature recognition scheme. Subsequently, the 3-D containment configuration is reconstructed from that of the 2-D by using a strategy calleddrop, pull and push. In this context, we introduce the notion of axis as a characteristic direction of a feature that facilitates the analysis of generalized 3-D polyhedral containment. The algorithms are applicable to generic B-Rep CAD data, and their time and space complexities are polynomial.     

Perceptual grouping of line features in 3-D space: a model-based framework

In this paper, we propose a novel model-based perceptual grouping algorithm for the line features of 3-D polyhedral objects. Given a 3-D polyhedral model, perceptual grouping is performed to extract a set of 3-D line segments which are geometrically consistent with the 3-D model. Unlike the conventional approaches, grouping is done in 3-D space in a model-based framework. In our unique approach, a decision tree classifier is employed for encoding and retrieving the geometric information of the 3-D model. A Gestalt graph is constructed by classifying input instances into proper Gestalt relations using the decision tree. The Gestalt graph is then decomposed into a few subgraphs, yielding appropriate groups of features. As an application, we suggest a 3-D object recognition system which can be accomplished by selecting a best-matched group. In order to evaluate the performance of the proposed algorithm, experiments are carried out on both synthetic and real scenes.     

A Spatial Access-Oriented Implementation of a 3-D GIS Topological Data Model for Urban Entities

3-D analysis in GIS is still one of the most challenging topics for research. With the goal being to model possible movement within the built environment, this paper, therefore, proposes a new approach to handling connectivity relationships among 3-D objects in urban environments in order to implement spatial access analyses in 3-D space. To achieve this goal, this paper introduces a 3-D network data model called the geometric network model (GNM), which has been developed by transforming the combinatorial data model (CDM), representing a connectivity relationship among 3-D objects using a dual graph. For the transformation, this paper presents (1) an O(n ²) algorithm for computing a straight medial axis transformation (MAT), (2) the processes for transforming phenomena from 3-D CDM to 3-D GNM, and (3) spatial access algorithms for the 3-D geometric network based upon the Dijkstra algorithm. Using the reconstructed geometric network generated from the transformations, spatial queries based upon the complex connectivity relationships between 3-D urban entities are implemented using Dijkstra algorithm. Finally, the paper presents the results of an experimental implementation of a 3-D network data model (GNM) using GIS data of an area in downtown Columbus, Ohio.     

Real-time part position sensing

A light-stripe vision system is used to measure the location of polyhedral features of parts from a single frame of video camera output. The geometric conditions which assure location of the feature when the light plane intersects three of the feature's faces are given. Issues such as accuracy in locating the line segments of intersection in the image and combining redundant information from multiple measurements and multiple sources are addressed. It was found that in 2.5 s, the prototype sensor was capable of locating a 2-in cube to an accuracy (one standard deviation) of 0.002 in (0.055 mm) in translation and 0.1° (0.0015 radians) in rotation. When integrated with a manipulator, the system was capable of performing high-precision assembly tasks     

Computation of force-closure grasps: an iterative algorithm

Computation of grasps with form/force closure is one of the fundamental problems in the study of multifingered grasping and dextrous manipulation. Based on the geometric condition of the closure property, this paper presents a numerical test to quantify how far a grasp is from losing form/force closure. With the polyhedral approximation of the friction cone, the proposed numerical test can be formulated as a single linear program. An iterative algorithm for computing optimal force-closure grasps, which is implemented by minimizing the proposed numerical test in the grasp configuration space, is also developed. The algorithm is computationally efficient and generally applicable. It can be used for computing form/force-closure grasps on 3-D objects with curved surfaces, and with any number of contact points. Simulation examples are given to show the effectiveness and computational efficiency of the proposed algorithm.     

Towards functional labeling of utility vehicle point clouds for humanoid driving

We present work on analyzing 3-D point clouds of a small utility vehicle for purposes of humanoid robot driving. The scope of this work is limited to a subset of ingress-related tasks including stepping up into the vehicle and grasping the steering wheel. First, we describe how partial point clouds are acquired from different perspectives using sensors, including a stereo camera and a tilting laser range finder. For finer detail and a larger model than one sensor view alone can capture, a Kinect Fusion (Izadi et al. in KinectFusion: realtime 3D reconstruction and interaction using a moving depth camera, 2011)-like algorithm is used to integrate the stereo point clouds as the sensor head is moved around the vehicle. Second, we discuss how individual sensor views can be registered to the overall vehicle model to provide context, and present methods to estimate both statically and dynamically several geometric parameters critical to motion planning: (1) the floor height and boundaries defined by the seat and the dashboard, and (2) the steering wheel pose and dimensions. Results are compared using the different sensors, and the usefulness of the estimated quantities for motion planning is also addressed.     

Near laser-scan quality 3-D face reconstruction from a low-quality depth stream

We propose a method to produce near laser-scan quality 3-D face models of a freely moving user with a low-cost, low resolution range sensor in real-time. Our approach does not require any prior knowledge about the geometry of a face and can produce faithful geometric models of any star-shaped object. We use a cylindrical representation, which enables us to efficiently process the 3-D mesh by applying 2-D filters.We use the first frame as a reference and incrementally build the model by registering each subsequent cloud of 3-D points to the reference using the ICP (Iterative Closest Point) algorithm implemented on a GPU (Graphics Processing Unit). The registered point clouds are merged into a single image through a cylindrical representation. The noise from the sensor and from the pose estimation error is removed with a temporal integration and a spatial smoothing of the successively incremented model. To validate our approach, we quantitatively compare our model to laser scans, and show comparable accuracy.¹     

Force and acceleration sensor fusion for compliant manipulation control in 6 degrees of freedom

This paper focusses on sensor fusion in robotic manipulation: six-dimensional (6-D) force/torque signals and 6-D acceleration signals are used to extract forces and torques caused by inertia. As result, only forces and torques established by environmental contact(s) remain. Apart from an improvement of hybrid force/pose control behavior, an additional major benefit is that regular resetting/zeroing of force/torque sensors before free space/contact transitions can be omitted. All essential equations, transformations and calculations that are required for this 6-D fusion approach are derived. To highlight the meaning for practical implementations, numerous experiments with a six-joint Staeubli RX60 industrial manipulator are presented and the achieved results are discussed.     

对未知环境的机器人力控自律跟踪及建模

在考虑摩擦力的情况下，利用传感器所感知的机器人和未知轮廓间的相互作用力确接触处轮廓的切矢和法矢，据此建立端点约束坐标系，在该坐标系中沿切矢进行位置控制并沿法矢进行力控制。实现机器人对未知轮廓的自律跟踪运动。由跟踪运动所确定每一点处切矢信息及该点位置信息构造未知轮廓几何模型。在机器人学开放研究实验室的ＰＵＭＡ５６２机器人上实现了上述自律运动并建立了环境模型。     

Personal authentication using 3-D finger geometry

In this paper, a biometric authentication system based on measurements of the user's three-dimensional (3-D) hand geometry is proposed. The system relies on a novel real-time and low-cost 3-D sensor that generates a dense range image of the scene. By exploiting 3-D information we are able to limit the constraints usually posed on the environment and the placement of the hand, and this greatly contributes to the unobtrusiveness of the system. Efficient, close to real-time algorithms for hand segmentation, localization and 3-D feature measurement are described and tested on an image database simulating a variety of working conditions. The performance of the system is shown to be similar to state-of-the-art hand geometry authentication techniques but without sacrificing the convenience of the user.     

基于顶点聚类的多面体模型简化方法

本文介绍一种基于顶点聚类的多面体模型简化方法 ,该方法主要由三部分组成 :( 1 )把多面体模型划分成若干个小单元 ;( 2 )落在同一单元中的一组网格顶点用一个代表顶点表示 ;( 3)由代表顶点进行消去操作 ,得到简化的三角形网格模型。该方法实现简单、速度快 ,并且具有通用性。     

Enhanced reliable reactive routing (ER3) protocol for multimedia applications in 3D wireless sensor networks

Sensor networks designed especially for the multimedia applications require high data rate and better Quality of Service (QoS). Offering a reliable and energy efficient routing technique in a harsh and complex three-dimensional (3-D) environment for multimedia applications is a challenging job. Geo-routing and geometric routing have been efficient routing schemes for two-dimensional (2-D), but are unable to work properly for 3-D sensor networks. In order to enhance the resilience to link the dynamics in the 3-D sensor network, in this research an Enhanced Reliable Reactive Routing (ER3) is proposed. ER3 is an advancement to the existing reactive routing schemes, to provide energy efficient and reliable routing of data packets in the complex 3-D sensor networks for multimedia applications. The major attraction of ER3 is its backoff scheme, which occurs in the route discovery phase. In backoff scheme robust pilot paths formed between the source and destination are calculated to enable cooperative forwarding of the data packets. The data packets in ER3 are forwarded greedily to the destination from the source and doesn’t require any prior location information of the nodes. The encompassing simulations suggest that the ER3 outperforms the existing routing protocols on the basis of energy efficiency, low latency and high packet delivery ratio.

     

A Polyhedral Bound on the Indeterminate Contact Forces in Planar Quasi-Rigid Fixturing and Grasping Arrangements

This paper considers multiple-contact arrangements where several bodies grasp, fixture, or support an object via frictional point contacts. Within a strictly rigid-body modeling paradigm, when an external wrench (i.e., force and torque) acts on the object, the reaction forces at the contacts are typically indeterminate and span an unbounded linear space. This paper analyzes the contact reaction forces within a generalized quasi-rigid-body framework that keeps the desirable geometric properties of rigid-body modeling, while also including more realistic physical effects. We describe two basic principles that govern the contact mechanics of quasi-rigid bodies. The main result is that for any given external wrench acting on a quasi-rigid object, the statically feasible contact reaction forces lie in a bounded polyhedral set that depends on the external wrench, the grasp's geometry, and the preload forces. Moreover, the bound does not depend upon any detailed knowledge of the contact mechanics parameters. When some knowledge of the parameters is available, the bound can be sharpened. The polyhedral bound is useful for “robust” grasp and fixture synthesis. Given a set of external wrenches that may act upon an object, the grasp's geometry and preload forces can be chosen such that all of these external wrenches would be automatically supported by the contacts.     

Force modeling for tooth preparation in a dental training system

Feedback force is very important for novices to simulate tooth preparation by using the haptic interaction system (dental training system) in a virtual environment. In the process of haptic simulation, the fidelity of generated forces by a haptic device decides whether the simulation is successful. A force model computes feedback force, and we present an analytical force model to compute the force between a tooth and a dental pin during tooth preparation. The force between a tooth and a dental pin is modeled in two parts: (1) force to resist human’s operation and (2) friction to resist the rotation of the dental engine. The force to resist the human’s operation is divided into three parts in the coordinates that are constructed on the bottom center of the dental pin. In addition, we also consider the effects of dental-pin type, tooth stiffness, and contact geometry in the force model. To determine the parameters of the force model, we construct a measuring system by using machine vision and a force/torque sensor to track the human’s operations and measure the forces between the dental pins and teeth. Based on the measuring results, we construct the relation between the force and the human’s operation. The force model is implemented in the prototype of a dental training system that uses the Phantom as the haptic interface. Dentists performing virtual operations have confirmed the fidelity of feedback force.     

Simulation of dynamics of interacting rigid bodies including friction II: Software system design and implementation

This is the second of two papers that deal with the problem of modeling contact (impact, sliding, rolling) between unconstrained rigid bodies, including friction. In a companion paper [1] we showed that the main underlying problem concerns the ability to do efficient contact mechanics when bodies interact through impact and/or sustained contact. Contact mechanics involves two aspects: detection of contact between bodies and estimation of contact forces. These forces are complicated in character and difficult to estimate because they depend on the material response of the contacting objects, on the duration of contact (very short duration impact, or more sustained contact), frictional interaction at the surfaces, geometry of contact, etc. In [1] we proposed a conceptual model in which linear elastic (springs) and viscous (dampers) elements acting at points of contact between objects generate all contact forces. In this paper we describe how the contact model has been implemented in the software of a working computer simulation system. The major aspects of this process are: formulation of a discrete version of the contact model; calculation of model parameters to reflect material properties; geometric representation of objects (in our system, objects are modeled as convex polyhedra); algorithms to detect and evaluate contacts among objects (a process called contact analysis); and estimation and control of model response for stable numerical integration of equations of motion. A graphical user interface displays a three-dimensional (3-D) perspective animation of the solution using full color shaded surface images. While the simulation may not be accomplished in real time, solutions can be saved in files for real-time visualization.Authors are listed in alphabetical order