Long-term knowledge acquisition using contextual information in a memory-inspired robot architecture

In this paper, we present a novel cognitive framework allowing a robot to form memories of relevant traits of its perceptions and to recall them when necessary. The framework is based on two main principles: on the one hand, we propose an architecture inspired by current knowledge in human memory organisation; on the other hand, we integrate such an architecture with the notion of context, which is used to modulate the knowledge acquisition process when consolidating memories and forming new ones, as well as with the notion of familiarity, which is employed to retrieve proper memories given relevant cues. Although much research has been carried out, which exploits Machine Learning approaches to provide robots with internal models of their environment (including objects and occurring events therein), we argue that such approaches may not be the right direction to follow if a long-term, continuous knowledge acquisition is to be achieved. As a case study scenario, we focus on both robot–environment and human–robot interaction processes. In case of robot–environment interaction, a robot performs pick and place movements using the objects in the workspace, at the same time observing their displacement on a table in front of it, and progressively forms memories defined as relevant cues (e.g. colour, shape or relative position) in a context-aware fashion. As far as human–robot interaction is concerned, the robot can recall specific snapshots representing past events using both sensory information and contextual cues upon request by humans.     

Ontology enhancing process for a situated and curiosity-driven robot

Nowadays, robots need to be able to interact with humans and objects in a flexible way and should be able to share the same knowledge (physical and social) of the human counterpart. Therefore, there is a need for a framework for expressing and sharing knowledge in a meaningful way by building the world model. In this paper, we propose a new framework for human–robot interaction using ontologies as powerful way of representing information which promote the sharing of meaningful knowledge between different objects. Furthermore, ontologies are powerful notions able to conceptualise the world in which the object such as Robot is situated. In this research, ontology is considered as improved solution to the grounding problem and enables interoperability between human and robot. The proposed system has been evaluated on a large number of test cases; results were very promising and support the implementation of the solution.     

Sensor Fault Tolerance Method by Using a Bayesian Network for Robot Behavior

This paper presents FTBN, a new framework that performs learning autonomous mobile robot behavior and fault tolerance simultaneously. For learning behavior in the presence of a robot sensor fault this framework uses a Bayesian network. In the proposed framework, sensor data are used to detect a faulty sensor. Fault isolation is accomplished by changing the Bayesian network structure using interpreted evidence from robot sensors. Experiments including both simulation and a real robot are performed for door-crossing behavior using prior knowledge and sensor data at several maps. This paper explains the learning behavior, optimal tracking, exprimental setup and structure of the proposed framework. The robot uses laser and sonar sensors for door-crossing behavior, such that each sensor can be corrupted during the behavior. Experimental results show FTBN leads to robust behavior in the presence of a sensor fault as well as performing better compared to the conventional Bayesian method.     

采用自然语言的移动机器人任务编程

本文在机器人具备基本运动技能的基础上［1］，采用基于指令教导的学习方法．通 过自然语言教会机器人完成抽象化任务，并以程序体方式保存所学知识，也即通过自然语言 对话自动生成程序流．通过让机器人完成导航等任务，验证所提自然语言编程方法的可行性 ．     

Robot assembly planning using a task grammar augmented with heuristic knowledge

Automated assembly planning in a manufacturing environment requires not only mathematically sound formal methods and algorithmic computations but also heuristic knowledge. Much of this knowledge can be extracted from manual task manipulation strategies, such as the motion classification scheme used in methods-time measurement (MTM) studies. In this paper, we delineate various task-level operations in the context of robotic assembly, and show how these operations can be organized in the form of a task grammar. The proposed task grammar captures the intrinsic principle on how the sequence of robot operations should be ordered and how one high-level operation can be effectively decomposed into low-level operations. In order to control the process of robot task decomposition, we explicitly represent and apply qualitative heuristic knowledge about task constraints and operation applicability. In the paper, we first describe how syntactical knowledge about robot operations can be formulated for assembly-related manipulation tasks. Next, through illustrative examples, we attempt to show how qualitative knowledge can be effectively used in the task decomposition in three distinct ways: heuristic-based operation pattern matching, spatial-feature-based qualitative state envisionment, and canonical transformation of task environments.     

Three-Dimensional Path Planning of a Climbing Robot Using Mixed Integer Linear Programming

The City-Climber robot is a novel wall-climbing robot developed at The City College of New York that has the capability to move on floors, climb walls, walk on ceilings and transit between them. In this paper, we first develop the dynamic model of the City-Climber robot when it travel on different surfaces, i.e., floors, walls and ceilings, respectively. Then, we present a path planning method for the City-Climber robot using mixed integer linear programming (MILP) in three-dimensional (3-D) building environments that consist of objects with primitive geometrical shapes. MILP provides an optimization framework that can directly incorporate dynamic constraints with logical constraints such as obstacle avoidance and waypoint selection. In order to use MILP to solve the obstacle avoidance problem, we simplify and decouple the robot dynamic model into a linear system by introducing a restricting admissible controller. The decoupled model and obstacle can be rewritten as a linear program with mixed-integer linear constraints that account for the collision avoidance. A key benefit of this approach is that the path optimization can be readily solved using the AMPL and CPLEX optimization software with a MATLAB interface. Simulation results show that the framework of MILP is well suited for path planning and obstacle avoidance problems for the wall-climbing robot in 3-D environments.     

A framework for robotic clothing assistance by imitation learning

ABSTRACT

The recent demographic trend across developed nations shows a dramatic increase in the aging population, fallen fertility rates and a shortage of caregivers. Hence, the demand for service robots to assist with dressing which is an essential Activity of Daily Living (ADL) is increasing rapidly. Robotic Clothing Assistance is a challenging task since the robot has to deal with two demanding tasks simultaneously, (a) non-rigid and highly flexible cloth manipulation and (b) safe human–robot interaction while assisting humans whose posture may vary during the task. On the other hand, humans can deal with these tasks rather easily. In this paper, we propose a framework for robotic clothing assistance by imitation learning from a human demonstration to a compliant dual-arm robot. In this framework, we divide the dressing task into three phases, i.e. reaching phase, arm dressing phase, and body dressing phase. We model the arm dressing phase as a global trajectory modification using Dynamic Movement Primitives (DMP), while we model the body dressing phase toward a local trajectory modification applying Bayesian Gaussian Process Latent Variable Model (BGPLVM). We show that the proposed framework developed towards assisting the elderly is generalizable to various people and successfully performs a sleeveless shirt dressing task. We also present participants feedback on public demonstration at the International Robot Exhibition (iREX) 2017. To our knowledge, this is the first work performing a full dressing of a sleeveless shirt on a human subject with a humanoid robot.     

THE SPATIAL SEMANTIC HIERARCHY IMPLEMENTED WITH AN OMNIDIRECTIONAL VISION SYSTEM

In this article, we propose a new approach to the map building task: the implementation of the Spatial Semantic Hierarchy (SSH), proposed by B. Kuipers, on a real robot fitted with an omnidirectional camera. The original Kuiper's formulation of the SSH was slightly modified, in order to manage in a more efficient way the knowledge the real robot collects while moving in the environment. The sensory data experienced by the robot are transformed by the different levels of the SSH in order to obtain a compact representation of the environment. This knowledge is stored in the form of a topological map and, eventually, of a metrical map. The aim of this article is to show that a catadioptric omnidirectional camera is a good sensor for the SSH and nicely couples with several elements of the SSH. The panoramic view and rotational invariance of our omnidirectional camera makes the identification and labelling of places a simple matter. A deeper insight is that the tracking and identification of events on an omnidirectional image such as occlusions and alignments can be used for the segmentation of continuous sensory image data into the discrete topological and metric elements of a map. The proposed combination of the SSH and omnidirectional vision provides a powerful general framework for robot maping and offers new insights into the concept of “place.” Some preliminary experiments performed with a real robot in an unmodified office environment are presented.     

Transferring knowledge across robots: A risk sensitive approach

One of the most impressive characteristics of human perception is its domain adaptation capability. Humans can recognize objects and places simply by transferring knowledge from their past experience. Inspired by that, current research in robotics is addressing a great challenge: building robots able to sense and interpret the surrounding world by reusing information previously collected, gathered by other robots or obtained from the web. But, how can a robot automatically understand what is useful among a large amount of information and perform knowledge transfer? In this paper we address the domain adaptation problem in the context of visual place recognition. We consider the scenario where a robot equipped with a monocular camera explores a new environment. In this situation traditional approaches based on supervised learning perform poorly, as no annotated data are provided in the new environment and the models learned from data collected in other places are inappropriate due to the large variability of visual information. To overcome these problems we introduce a novel transfer learning approach. With our algorithm the robot is given only some training data (annotated images collected in different environments by other robots) and is able to decide whether, and how much, this knowledge is useful in the current scenario. At the base of our approach there is a transfer risk measure which quantifies the similarity between the given and the new visual data. To improve the performance, we also extend our framework to take into account multiple visual cues. Our experiments on three publicly available datasets demonstrate the effectiveness of the proposed approach.     

New evaluation framework for human-assistive devices based on humanoid robotics

This paper presents the new application of a humanoid robot as an evaluator of human-assistive devices. The reliable and objective evaluation framework for assistive devices is necessary for making industrial standards in order that those devices are used in various applications. In this framework, we utilize a recent humanoid robot with its high similarity to humans, human motion retargeting techniques to a humanoid robot, and identification techniques of robot’s mechanical properties. We also show two approaches to estimate supporting torques from the sensor data, which can be used properly according to the situations. With the general formulation of the wire-driven multi-body system, the supporting torque of passive assistive devices is also formulated. We evaluate a passive assistive wear ‘Smart Suit Lite (SSL)’ as an example of device, and use HRP-4 as the humanoid platform.     

On Intelligent Robot Motion Planning via Learning

Automatic motion planning is one of the basic modules that are needed to increase robot intelligence and usability. Unfortunately, the inherent complexity of motion planning has rendered traditional search algorithms incapable of solving every problem in real time. To circumvent this difficulty, we explore the alternative of allowing human operators to participate in the problem solving process. By having the human operator teach during difficult motion planning episodes, the robot should be able to learn and improve its own motion planning capability and gradually reduce its reliance on the human operator. In this paper, we present such a learning framework in which both human and robot can cooperate to achieve real-time automatic motion planning. To enable a deeper understanding of the framework in terms of performance, we present it as a simple learning algorithm and provide theoretical analysis of its behavior. In particular, we characterize the situations in which learning is useful, and provide quantitative bounds to predict the necessary training time and the maximum achievable speedup in planning time.     

Robotic grasping and manipulation through human visuomotor learning

A major goal of robotics research is to develop techniques that allow non-experts to teach robots dexterous skills. In this paper, we report our progress on the development of a framework which exploits human sensorimotor learning capability to address this aim. The idea is to place the human operator in the robot control loop where he/she can intuitively control the robot, and by practice, learn to perform the target task with the robot. Subsequently, by analyzing the robot control obtained by the human, it is possible to design a controller that allows the robot to autonomously perform the task. First, we introduce this framework with the ball-swapping task where a robot hand has to swap the position of the balls without dropping them, and present new analyses investigating the intrinsic dimension of the ball-swapping skill obtained through this framework. Then, we present new experiments toward obtaining an autonomous grasp controller on an anthropomorphic robot. In the experiments, the operator directly controls the (simulated) robot using visual feedback to achieve robust grasping with the robot. The data collected is then analyzed for inferring the grasping strategy discovered by the human operator. Finally, a method to generalize grasping actions using the collected data is presented, which allows the robot to autonomously generate grasping actions for different orientations of the target object.     

Hierarchically self-organizing visual place memory

A hierarchically organized visual place memory enables a robot to associate with its respective knowledge efficiently. In this paper, we consider how this organization can be done by the robot on its own throughout its operation and introduce an approach that is based on the agglomerative method SLINK. The hierarchy is obtained from a single link cluster analysis that is carried out based on similarity in the appearance space. As such, the robot can incrementally incorporate the knowledge of places into its visual place memory over the long term. The resulting place memory has an order-invariant hierarchy that enables both storage and construction efficiency. Experimental results obtained under the guided operation of the robot demonstrate that the robot is able to organize its place knowledge and relate to it efficiently. This is followed by experimental results under autonomous operation in which the robot evolves its visual place memory completely on its own.     

On-line motion synthesis and adaptation using a trajectory database

Autonomous robots cannot be programmed in advance for all possible situations. Instead, they should be able to generalize the previously acquired knowledge to operate in new situations as they arise. A possible solution to the problem of generalization is to apply statistical methods that can generate useful robot responses in situations for which the robot has not been specifically instructed how to respond. In this paper we propose a methodology for the statistical generalization of the available sensorimotor knowledge in real-time. Example trajectories are generalized by applying Gaussian process regression, using the parameters describing a task as query points into the trajectory database. We show on real-world tasks that the proposed methodology can be integrated into a sensory feedback loop, where the generalization algorithm is applied in real-time to adapt robot motion to the perceived changes of the external world.     

基于ROS的云机器人服务框架

随着机器人技术的发展和硬件的普及,机器人的市场应用前景越来越广泛.但是,目前的机器人仍存在着许多局限,具体表现为可移植性弱,本地计算成本高,服务应用较少等.云机器人的提出,将机器人的计算能力从本地迁移到云端,不但提高了机器人的计算能力,降低了机器人硬件成本,而且能使资源的分配更为均衡,为解决机器人发展面临的困难提供了有效的解决途径.在云机器人的基础思想之上,提出一种基于ROS的云机器人服务框架.该服务框架使用开源的机器人操作系统ROS作为机器人运行的基础,增强了其对于不同硬件和软件环境的可移植性.同时,在框架的云端部分加入了机器人的服务管理系统和服务解析模块,能够方便快速地对机器人服务进行扩展和调用.在最后的实验部分,通过人脸识别服务模块对云机器人服务框架进行了实验验证.     

Robot motion design using bunraku emotional expressions – focusing on Jo-Ha-Kyū in sounds and movements*

One of the UNESCO intangible cultural heritages Bunraku puppets can play one of the most beautiful puppet motions in the world. The Bunraku puppet motions can express emotions without the so-called ‘Uncanny Valley.’ We try to convert these emotional motions into robot affective motions so that robots can interact with human beings more comfortable. In so doing, in the present paper, we present a robot motion design framework using Bunraku affective motions that are based on the so-called ‘Jo-Ha-Kyū,’ and convert a few simple Bunraku motions into a robot motions using one of deep learning methods. Our primitive experiments show that Jo-Ha-Kyū can be incorporated into robot motion design smoothly, and some simple affective robot motions can be designed using our proposed framework.     

A constraint-based approach for human–robot collision avoidance

In this paper, we present a software-based approach for collision avoidance that can be applied in human–robot collaboration scenarios. One of the contributions is a method for converting clustered 3D sensor data into computationally efficient convex hull representations used for robot-obstacle distance computation. Based on the computed distance vectors, we generate collision avoidance motions using a potential field approach and integrate them with other simultaneously running robot tasks in a constraint-based control framework. In order to improve control performance, we apply evolutionary techniques for parameter optimization within this framework based on selected quality criteria. Experiments are performed on a dual-arm robotic system equipped with several depth cameras. The approach is able to generate task-compliant avoidance motions in dynamic environments with high performance.     

Active object tracking using context estimation: handling occlusions and detecting missing targets

When performing visual servoing or object tracking tasks, active sensor planning is essential to keep targets in sight or to relocate them when missing. In particular, when dealing with a known target missing from the sensor’s field of view, we propose using prior knowledge related to contextual information to estimate its possible location. To this end, this study proposes a Dynamic Bayesian Network that uses contextual information to effectively search for targets. Monte Carlo particle filtering is employed to approximate the posterior probability of the target’s state, from which uncertainty is defined. We define the robot’s utility function via information theoretic formalism as seeking the optimal action which reduces uncertainty of a task, prompting robot agents to investigate the location where the target most likely might exist. Using a context state model, we design the agent’s high-level decision framework using a Partially-Observable Markov Decision Process. Based on the estimated belief state of the context via sequential observations, the robot’s navigation actions are determined to conduct exploratory and detection tasks. By using this multi-modal context model, our agent can effectively handle basic dynamic events, such as obstruction of targets or their absence from the field of view. We implement and demonstrate these capabilities on a mobile robot in real-time.

     

Multi-rigid-body dynamics and online model predictive control for transformable multi-links aerial robot*

Transformable multi-links aerial robots have great potentials in application relying on the transformable features to change its shape during the flight. Compared to traditional quadrotor robots, transformable multi-links robots are equipped with servo motor between links. To simplify the non-linear dynamic system, the previous work restricts the robot to transform in very slow speed so that the robot could be approximated as a quadrotor robot at each time point. However, tradeoff comes as the dynamic performance is given up. In this paper, we come up with a new framework combining of computationally efficient non-linear model predictive controller and motion primitive to optimize thrust force and joints trajectory of the multi-links aerial robot. Finally, we verify our framework with fast transformation motions and table tennis task which requires dynamic performance.