nach oben

2014 | Buch

Kapitel lesen Erstes Kapitel lesen

Advances in Autonomous Robotics Systems

15th Annual Conference, TAROS 2014, Birmingham, UK, September 1-3, 2014. Proceedings

herausgegeben von: Michael Mistry, Aleš Leonardis, Mark Witkowski, Chris Melhuish

Verlag: Springer International Publishing

Buchreihe : Lecture Notes in Computer Science

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten

Über dieses Buch

This book constitutes the refereed proceedings of the 15th Conference on Advances in Autonomous Robotics, TAROS 2014, held in Birmingham, UK, in September 2014. The 23 revised full papers presented together with 9 extended abstracts were carefully reviewed and selected from 48 submissions. The overall program covers various aspects of robotics, including navigation, planning, sensing and perception, flying and swarm robots, ethics, humanoid robotics, human-robot interaction, and social robotics.

Inhaltsverzeichnis

Frontmatter

Full Papers

Modeling of a Large Structured Environment

With a Repetitive Canonical Geometric-Semantic Model

Abstract

AIMS project attempts to link the logistic requirements of an intelligent warehouse and state of the art core technologies of automation, by providing an awareness of the environment to the autonomous systems and vice versa. In this work we investigate a solution for modeling the infrastructure of a structured environment such as warehouses, by the means of a vision sensor. The model is based on the expected pattern of the infrastructure, generated from and matched to the map. Generation of the model is based on a set of tools such as closed-form Hough transform, DBSCAN clustering algorithm, Fourier transform and optimization techniques. The performance evaluation of the proposed method is accompanied with a real world experiment.

Saeed Gholami Shahbandi, Björn Åstrand

Monte Carlo Localization for Teach-and-Repeat Feature-Based Navigation

Abstract

This work presents a combination of a teach-and-replay visual navigation and Monte Carlo localization methods. It improves a reliable teach-and-replay navigation method by replacing its dependency on precise dead-reckoning by introducing Monte Carlo localization to determine robot position along the learned path. In consequence, the navigation method becomes robust to dead-reckoning errors, can be started from at any point in the map and can deal with the ‘kidnapped robot’ problem. Furthermore, the robot is localized with MCL only along the taught path, i.e. in one dimension, which does not require a high number of particles and significantly reduces the computational cost. Thus, the combination of MCL and teach-and-replay navigation mitigates the disadvantages of both methods. The method was tested using a P3-AT ground robot and a Parrot AR.Drone aerial robot over a long indoor corridor. Experiments show the validity of the approach and establish a solid base for continuing this work.

Matías Nitsche, Taihú Pire, Tomáš Krajník, Miroslav Kulich, Marta Mejail

An Improved Cellular Automata-Based Model for Robot Path-Planning

Abstract

Cellular automata (CA) are able to represent high complex phenomena and can be naturally simulated by digital processors due to its intrinsic discrete nature. CA have been recently considered for path planning in autonomous robotics. In this work we started by adapting a model proposed by Ioannidis et al. to deal with scenarios with a single robot, turning it in a more decentralized approach. However, by simulating this model we noticed a problem that prevents the robot to continue on its path and avoid obstacles. A new version of the model was then proposed to solve it. This new model uses CA transition rules with Moore neighborhood and four possible states per cell. Simulations and experiments involving real e-puck robots were performed to evaluate the model. The results show a real improvement in the robot performance.

Giordano B. S. Ferreira, Patricia A. Vargas, Gina M. B. Oliveira

Bioinspired Mechanisms and Sensorimotor Schemes for Flying: A Preliminary Study for a Robotic Bat

Abstract

The authors present a critical review on flying motion and echolocation in robotics, directly derived from research on biology and flight mechanics of bats, aimed at designing and prototyping a new generation of robotic technologies.

To achieve this goal, the paper starts with an in-depth review on real bats, studying and proposing models derived from the analysis of the aerodynamics of bat wings combined together with the investigation of the bat echolocation system.

These models will form the basis for the implementation of the conceptual design of the proposed robotic platform, consisting of a new complete bat body, light and smart body structures, provided with a set of sensors and actuated joints, displaying echolocation, sensing and coordination capabilities.

Carmine Tommaso Recchiuto, Rezia Molfino, Anders Hedenströem, Herbert Peremans, Vittorio Cipolla, Aldo Frediani, Emanuele Rizzo, Giovanni Gerardo Muscolo

Evolutionary Coordination System for Fixed-Wing Communications Unmanned Aerial Vehicles

Abstract

A system to coordinate the movement of a group of unmanned aerial vehicles that provide a network backbone over mobile ground-based vehicles with communication needs is presented. Using evolutionary algorithms, the system evolves flying manoeuvres that position the aerial vehicles by fulfilling two key requirements; i) they maximise net coverage and ii) they minimise the power consumption. Experimental results show that the proposed coordination system is able to offer a desirable level of adaptability with respect to the objectives set, providing useful feedback for future research directions.

Alexandros Giagkos, Elio Tuci, Myra S. Wilson, Philip B. Charlesworth

Multi-agent Environment Exploration with AR.Drones

Abstract

This paper describes work on a framework for multi-agent research using low cost Micro Aerial Vehicles (MAV’s). In the past this type of research has required significant investment for both the vehicles themselves and the infrastructure necessary to safely conduct experiments. We present an alternative solution using a robust, low cost, off the shelf platform. We demonstrate the capabilities of our system via two typical multi-robot tasks: obstacle avoidance and exploration. Developing multi-agent applications safely and quickly can be difficult using hardware alone, to address this we also present a multi-quadcopter simulation based around the Gazebo 3D simulator.

Richard Williams, Boris Konev, Frans Coenen

H ∞ Path Tracking Control for Quadrotors Based on Quaternion Representation

Abstract

In this work, the path tracking problem of quadrotors is investigated. A quadrotor is represented by unit quaternion and modeled with added disturbance. Given full and accurate location information, a nonlinear H _∞ control law is proposed and its stability is analyzed by using Lyapunov stability theorem. The added disturbance includes parameter changes and force disturbance. The simulation result demonstrates the closed loop path tracking system is stable with and without the added disturbance.

Wesam Jasim, Dongbing Gu

Towards an Ethical Robot: Internal Models, Consequences and Ethical Action Selection

Abstract

If robots are to be trusted, especially when interacting with humans, then they will need to be more than just safe. This paper explores the potential of robots capable of modelling and therefore predicting the consequences of both their own actions, and the actions of other dynamic actors in their environment. We show that with the addition of an ‘ethical’ action selection mechanism a robot can sometimes choose actions that compromise its own safety in order to prevent a second robot from coming to harm. An implementation with e-puck mobile robots provides a proof of principle by showing that a simple robot can, in real time, model and act upon the consequences of both its own and another robot’s actions. We argue that this work moves us towards robots that are ethical, as well as safe.

Alan F. T. Winfield, Christian Blum, Wenguo Liu

“The Fridge Door is Open”–Temporal Verification of a Robotic Assistant’s Behaviours

Abstract

Robotic assistants are being designed to help, or work with, humans in a variety of situations from assistance within domestic situations, through medical care, to industrial settings. Whilst robots have been used in industry for some time they are often limited in terms of their range of movement or range of tasks. A new generation of robotic assistants have more freedom to move, and are able to autonomously make decisions and decide between alternatives. For people to adopt such robots they will have to be shown to be both safe and trustworthy. In this paper we focus on formal verification of a set of rules that have been developed to control the Care-O-bot, a robotic assistant located in a typical domestic environment. In particular, we apply model-checking, an automated and exhaustive algorithmic technique, to check whether formal temporal properties are satisfied on all the possible behaviours of the system. We prove a number of properties relating to robot behaviours, their priority and interruptibility, helping to support both safety and trustworthiness of robot behaviours.

Clare Dixon, Matt Webster, Joe Saunders, Michael Fisher, Kerstin Dautenhahn

Implementation and Test of Human-Operated and Human-Like Adaptive Impedance Controls on Baxter Robot

Abstract

This paper presents an improved method to teleoperate impedance of a robot based on surface electromyography (EMG) and test it experimentally. Based on a linear mapping between EMG amplitude and stiffness, an incremental stiffness extraction method is developed, which uses instantaneous amplitude identified from EMG in a high frequency band, compensating for non-linear residual error in the linear mapping and preventing muscle fatigue from affecting the control. Experiments on one joint of the Baxter robot are carried out to test the approach in a disturbance attenuation task, and to compare it with automatic human-like impedance adaptation. The experimental results demonstrate that the new human operated impedance method is successful at attenuating disturbance, and results similarly to as automatic disturbance attenuation, thus demonstrating its efficiency.

Peidong Liang, Chenguang Yang, Ning Wang, Zhijun Li, Ruifeng Li, Etienne Burdet

Hybrid Communication System for Long Range Resource Tracking in Search and Rescue Scenarios

Abstract

This paper focuses on a resource (people or equipment) tracking approach involving radio localization and wireless communication for use in search and rescue scenarios. Two alternative communication options (ZigBee and GSM) are used to facilitate robust wireless data transfers between nodes through either an ad-hoc network or base station. The onboard sensors of the tracking device calculate the speed of the node’s movement as well as other environmental parameters such as temperature. The position of each node is regularly uploaded to a map on a server. XBee modules are used for local communication within a range of several hundred meters, while GSM is used for long range communication provided that the infrastructure is available. This paper discusses the hardware platform and network architecture of the developed hybrid communication system. Experiments involving device prototypes are performed to evaluate the feasibility and performance of the system. The results show that a hybrid communication system is extremely practical for search and rescue operations.

Michael Boor, Ulf Witkowski, Reza Zandian

Wearable Self Sufficient MFC Communication System Powered by Urine

Abstract

A new generation of self-sustainable and wearable Microbial Fuel Cells (MFCs) is introduced. Two different types of energy - chemical energy found in urine and mechanical energy harvested by manual pumping - were converted to electrical energy. The wearable system is fabricated using flexible MFCs with urine used as the feedstock for the bacteria, which was pumped by a manual foot pump. The pump was developed using check valves and soft tubing. The MFC system has been assembled within a pair of socks.

Majid Taghavi, Andrew Stinchcombe, John Greenman, Virgilio Mattoli, Lucia Beccai, Barbara Mazzolai, Chris Melhuish, Ioannis A. Ieropoulos

Morphogenetic Self-Organization of Collective Movement without Directional Sensing

Abstract

In this paper, we present a morphogenetic approach to self-organized collective movement of a swarm. We assume that the robots (agents) do not have global knowledge of the environment and can communicate only locally with other robots. In addition, we assume that the robots are not able to perform directional sensing. To self-organize such systems, we adopt here a simplified diffusion mechanism inspired from biological morphogenesis. A guidance mechanism is proposed based on the history of morphogen concentrations. The division of labor is achieved by type differentiation to allocate different tasks to different type of robots. Simulations are run to show the efficiency of the proposed algorithm. The robustness of the algorithm is demonstrated by introducing an obstacle into the environment and removing a subset of robots from the swarm.

Ataollah Ramezan Shirazi, Hyondong Oh, Yaochu Jin

The Pi Swarm: A Low-Cost Platform for Swarm Robotics Research and Education

Abstract

The paper introduces the Pi Swarm robot, a platform developed to allow research and education in swarm robotics. Motivated by the goals of reducing costs and simplifying the tool-chain and programming knowledge needed to investigate swarming algorithms, we have developed a trackable, sensor-rich and expandable platform which needs only a computer with internet browser and no additional software to program. This paper details the design and use of the robot in a variety of settings, and we feel the platform makes for a viable, low-cost alternative for development of swarm robotic solutions.

James Hilder, Rebecca Naylor, Artjoms Rizihs, Daniel Franks, Jon Timmis

Tactile Features: Recognising Touch Sensations with a Novel and Inexpensive Tactile Sensor

Abstract

A simple and cost effective new tactile sensor is presented, based on a camera capturing images of the shading of a deformable rubber membrane. In Computer Vision, the issue of information encoding and classification is well studied. In this paper we explore different ways of encoding tactile images, including: Hu moments, Zernike Moments, Principal Component Analysis (PCA), Zernike PCA, and vectorized scaling. These encodings are tested by performing tactile shape recognition using a number of supervised approaches (Nearest Neighbor, Artificial Neural Networks, Support Vector Machines, Naive Bayes). In conclusion: the most effective way of representing tactile information is achieved by combining Zernike Moments and PCA, and the most accurate classifier is Nearest Neighbor, with which the system achieves a high degree (96.4%) of accuracy at recognising seven basic shapes.

Tadeo Corradi, Peter Hall, Pejman Iravani

Polygonal Models for Clothing

Abstract

We address the problem of recognizing a configuration of a piece of garment fairly spread out on a flat surface. We suppose that the background surface is invariant and that its color is sufficiently dissimilar from the color of a piece of garment. This assumption enables quite reliable segmentation followed by extraction of the garment contour. The contour is approximated by a polygon which is then fitted to a polygonal garment model. The model is specific for each category of garment (e.g. towel, pants, shirt) and its parameters are learned from training data. The fitting procedure is based on minimization of the energy function expressing dissimilarities between observed and expected data. The fitted model provides reliable estimation of garment landmark points which can be utilized for an automated folding using a pair of robotic arms. The proposed method was experimentally verified on a dataset of images. It was also deployed to a robot and tested in a real-time automated folding.

Jan Stria, Daniel Průša, Václav Hlaváč

Design of a Multi-purpose Low-Cost Mobile Robot for Research and Education

Abstract

Mobile robots are commonly used for research and education. Although there are several commercial mobile robots available for these tasks, they are often costly, do not always meet the characteristics needed for certain applications and are very difficult to adapt because they have proprietary software and hardware. In this paper, we present the design principles, and describe the development and applications of a mobile robot called ExaBot. Our main goal was to obtain a single multi-purpose low-cost robot -more than ten times cheaper than commercially available platforms- that can be used not only for research, but also for education and public outreach activities. The body of the ExaBot, its sensors, actuators, processing units and control board are described in detail. The software and printed circuit board developed for this project are open source to allow the robotics community to use and upgrade the current version. Finally, different configurations of the ExaBot are presented, showing several applications that fulfill the requirements this robotic platform was designed for.

Sol Pedre, Matías Nitsche, Facundo Pessagc, Javier Caccavelli, Pablo De Cristóforis

Adaptive Swarm Robot Region Coverage Using Gene Regulatory Networks

Abstract

This paper proposes a morphogenetic pattern formation approach for collective systems to cover a desired region for target entrapment. This has been achieved by combining a two-layer hierarchical gene regulatory network (H-GRN) with a region-based shape control strategy. The upper layer of the H-GRN is for pattern generation that provides a desired region for entrapping targets generated from local sensory inputs of detected targets. This pattern is represented by a set of arc segments, which allow us to form entrapping shape constraints with the minimum information that can be easily used by the lower layer of the H-GRN. The lower layer is for region-based shape control consisting of two steps: guiding all robots into the desired region designated by the upper layer, and maintaining a specified minimum distance between each robot and its neighbouring robots. Numerical simulations have been performed for scenarios containing either static and moving targets to validate the feasibility and benefits of the proposed approach.

Hyondong Oh, Yaochu Jin

Communicating Unknown Objects to Robots through Pointing Gestures

Abstract

Delegating tasks from a human to a robot needs an efficient and easy-to-use communication pipeline between them - especially when inexperienced users are involved. This work presents a robotic system that is able to bridge this communication gap by exploiting 3D sensing for gesture recognition and real-time object segmentation. We visually extract an unknown object indicated by a human through a pointing gesture and thereby communicating the object of interest to the robot which can be used to perform a certain task. The robot uses RGB-D sensors to observe the human and find the 3D point indicated by the pointing gesture. This point is used to initialize a fixation-based, fast object segmentation algorithm, inferring thus the outline of the whole object. A series of experiments with different objects and pointing gestures show that both the recognition of the gesture, the extraction of the pointing direction in 3D, and the object segmentation perform robustly. The discussed system can provide the first step towards more complex tasks, such as object recognition, grasping or learning by demonstration with obvious value in both industrial and domestic settings.

Bjarne Großmann, Mikkel Rath Pedersen, Juris Klonovs, Dennis Herzog, Lazaros Nalpantidis, Volker Krüger

A Cost-Effective Automatic 3D Reconstruction Pipeline for Plants Using Multi-view Images

Abstract

Plant phenotyping involves the measurement, ideally objectively, of characteristics or traits. Traditionally, this is either limited to tedious and sparse manual measurements, often acquired destructively, or coarse image-based 2D measurements. 3D sensing technologies (3D laser scanning, structured light and digital photography) are increasingly incorporated into mass produced consumer goods and have the potential to automate the process, providing a cost-effective alternative to current commercial phenotyping platforms. We evaluate the performance, cost and practicability for plant phenotyping and present a 3D reconstruction method from multi-view images acquired with a domestic quality camera. This method consists of the following steps: (i) image acquisition using a digital camera and turntable; (ii) extraction of local invariant features and matching from overlapping image pairs; (iii) estimation of camera parameters and pose based on Structure from Motion(SFM); and (iv) employment of a patch based multi-view stereo technique to implement a dense 3D point cloud. We conclude that the proposed 3D reconstruction is a promising generalized technique for the non-destructive phenotyping of various plants during their whole growth cycles.

Lu Lou, Yonghuai Liu, Minglan Sheng, Jiwan Han, John H. Doonan

Improving the Generation of Rapidly Exploring Randomised Trees (RRTs) in Large Scale Virtual Environments Using Trails

Abstract

Rapidly exploring randomised trees (RRTs) are a useful tool generating maps for use by agents to navigate. A disadvantage to using RRTs is the length of time required to generate the map. In large scale environments, or those with narrow corridors, the time needed to create the map can be prohibitive. This paper explores a new method for improving the generation of RRTs in large scale environments. We look at using trails as a new source of information for the agent’s map building process. Trails are a set of observations of how other agents, human or AI, have navigated an environment. We evaluate RRT performance in two types of virtual environment, the first generated to cover a variety of scenarios an agent may face when building maps, the second is a set of ‘real’ virtual environments based in Second Life. By including trails we can improve the RRT generation step in most environments, allowing the RRT to be used to successfully plan routes using fewer points and reducing the length of the overall route.

Katrina Samperi, Nick Hawes

Intelligent Computation of Inverse Kinematics of a 5-dof Manipulator Using MLPNN

Abstract

This paper presents inverse kinematic solution of 5 degree of freedom robot manipulator. Inverse kinematics is computation of all joint angles and link geometries which could be used to reach the given position and orientation of the end effector. This computation is very difficult to attain exact solution for the position and orientation of end effector due to the nature of non- algebraic equation of inverse kinematics. Therefor it is required to use some soft computing technique for the solution of inverse kinematics of robot manipulator. This paper presents structured artificial neural network (ANN) model from soft computing domain. The ANN model used is a Multi Layered Perceptron Neural Network (MLPNN). In this gradient descent type of learning rules are applied. An attempt has been made to find the best ANN configuration for the problem. It was found that between multi-layered perceptron neural network giving better result and calculated mean square error, as the performance index.

Panchanand Jha, Bibhuti Bhusan Biswal, Om Prakash Sahu

Humanoid Robot Gait Generator: Foot Steps Calculation for Trajectory Following

Abstract

During bipedal gait, a robot falls from one foot to the other. This motion can be approximated with that of an inverted pendulum with discrete movements of the contact point. We detail here how to use the linear inverted pendulum model (LIPM) for selecting successive contact points in such a way that the centre of mass (COM) of the robot flexibly follows a predefined set of waypoints on a straight or curved trajectory, allowing it to move forward, stop and revert its direction of motion in stable way. The use of a fixed step cycle duration reduces the mathematical complexity and the computational load, enabling real-time updating of gait parameters in a microcontroller.

Horatio Garton, Guido Bugmann, Phil Culverhouse, Stephen Roberts, Claire Simpson, Aashish Santana

Erratum: H ∞ Path Tracking Control for Quadrotors Based on Quaternion Representation

Abstract

The address of the first author of the paper starting on page 72 of this volume is not complete. It should be:

Wesam Jasim^1,2 and Dongbing Gu¹

¹ School of Computer Science and Electronic Engineering, University of Essex, Wivenhoe Park, Colchester, UK.

² University of Alanbar, Alanbar, Iraq

Wesam Jasim, Dongbing Gu

Backmatter

Titel: Advances in Autonomous Robotics Systems
herausgegeben von: Michael Mistry
Aleš Leonardis
Mark Witkowski
Chris Melhuish
Verlag: Springer International Publishing
Electronic ISBN: 978-3-319-10401-0
Print ISBN: 978-3-319-10400-3
DOI: https://doi.org/10.1007/978-3-319-10401-0