Please wait a minute...
投稿  |   English  | 
 
高级检索
   首页  |  最新收录  |  当期目录  |  过刊浏览  |  作者中心  |  关于期刊   开放获取  
投稿  |   English  | 
Engineering    2015, Vol. 1 Issue (1) : 27-35     https://doi.org/10.15302/J-ENG-2015007
研究型文章 |
助老服务机器人系统设计及软件架构
Hendrich Norman(),Bistry Hannes,张建伟
Computer Science Department, University of Hamburg, Hamburg D-22527, Germany
全文: PDF(8114 KB)   HTML
导出: BibTeX | EndNote | Reference Manager | ProCite | RefWorks     支持信息
文章导读  
摘要 

将智能人居辅助环境系统与服务机器人技术相结合,可以有效帮助老年人进行很多日常活动,有利于老年人获得更加良好的生活状态。本文概述了欧盟项目Robot-Era开发的智能人居辅助环境(AAL)系统,并重点阐述了系统中具有室内移动及物品操作能力的机器人的工程实现方法以及软件架构。该系统基于机器人操作系统(ROS)对大量先进的导航定位、环境感知以及操作控制算法进行集成,并通过实验对机器人的性能和实际应用效果进行验证。

关键词 服务机器人人居辅助环境操作及抓取用户体验分析    
Abstract

Systems for ambient assisted living (AAL) that integrate service robots with sensor networks and user monitoring can help elderly people with their daily activities, allowing them to stay in their homes and live active lives for as long as possible. In this paper, we outline the AAL system currently developed in the European project Robot-Era, and describe the engineering aspects and the service-oriented software architecture of the domestic robot, a service robot with advanced manipulation capabilities. Based on the robot operating system (ROS) middleware, our software integrates a large set of advanced algorithms for navigation, perception, and manipulation. In tests with real end users, the performance and acceptability of the platform are evaluated.

Keywords service robots      ambient assisted living      manipulation and grasping      user study     
基金资助: 
通讯作者: Hendrich Norman     E-mail: hendrich@informatik.uni-hamburg.de
最新录用日期:    发布日期: 2015-07-02
服务
推荐给朋友
免费邮件订阅
RSS订阅
作者相关文章
Norman Hendrich
Hannes Bistry
Jianwei Zhang
引用本文:   
Norman Hendrich,Hannes Bistry,Jianwei Zhang. Architecture and Software Design for a Service Robot in an Elderly-Care Scenario[J]. Engineering, 2015, 1(1): 27-35.
网址:  
http://engineering.ckcest.cn/eng/EN/10.15302/J-ENG-2015007     OR     http://engineering.ckcest.cn/eng/EN/Y2015/V1/I1/27
Fig.1  The Robot-Era domestic robot. (a) The robot combines the ScitosG5 differential-drive platform, the Kinova Jaco 6-DOF arm with integrated three-finger hand, and a pan-tilt sensor head; (b) close-ups of the tilting handle for walking support; (c) the object transportation tray.
Fig.2  Overall concept and architecture of the Robot-Era platform for ambient assisted living and elderly care. The end users and caregivers interact with the system using a speech interface or an intentionally simple tablet-based graphical user interface. Different robots are used for outdoor, condominium (transport), and domestic (cleaning) uses. Data from ambient sensors, user monitoring sensors, the robots, and smart appliances is integrated in the PEIS middleware layer. The configuration planner coordinates the system and the robots.
Fig.3  Evolution of the domestic robot. (a) Initial design sketch and concept of the tilting handle for walking support; (b) experimental Jaco arm integration on the astromobile robot [19]; (c) ROS URDF robot model with sensor and actuator coordinate systems; (d) first-year prototype without cover; (e) second-year prototype with soft cover; (f) third-year prototype with hardware updates and new cover; (g) experimental test.
Fig.4  Robot arms evaluated for the domestic robot due to their combination of weight, payload, reach, mobile use, and costs. (a) Schunk Powerball arm; (b) Kinova Jaco; (c) close-up of the Jaco three-finger hand; (d) Schunk modular arm; (e) BionicRobots BioRob arm (memory-wire actuation); (f) Universal Robots UR5. (Images courtesy of the vendors)
Fig.5  Example workspace analysis of the robot with the Jaco arm in top-grasp hand orientation. The arm-mount position was chosen so that the robot can reach the floor and access its object transportation tray, and so that it has a large workspace on its right side. (a) Reaching the floor; (b) example from a hospital-care scenario.
Fig.6  Software architecture for the domestic robot. Users request robot services from the PEIS middleware using speech or their tablet computers (HRI). PEIS manages the ambient sensor network and provides the symbolic planner. The PEIS-ROS exekutor modules encapsulate the robot services; this layer is subdivided into high-, medium-, and low-level robot skills (layer 2). The main robot software (layer 3) is fully based on ROS, except for platform localization and navigation handled by MIRA/CogniDrive. Layer 4 includes the device drivers that control the sensors and actuators.
SkillLevelDescription
Emergency StopLSafe stop of all robot motion
GetCameraImageLSend image from camera to PEIS
GetKinectImageLGets image and 3D pointcloud
GetLaserScanLReads laser scan for navigation/docking
MoveToLDrives the robot to the given pose (x, y, q)
MovePtuLMoves the robot head to given (pan, tilt)
RetractJacoArmLMoves the arm to the save park position
DetectKnownObjectIFind the requested object, return 6-DOF position (x, y, z) and orientation (j, y, q)
GraspKnownObjectIMove the hand to the object and grasp it
PlaceObjectOnTrayIPuts a grasped object onto the robot’s tray
MoveHingedDoorIArm motion to grasp and open a door
DockToCoroIDrive to position of the condominium robot, align robots for object exchange
HandoverObjectIMove the arm toward the user, wait for voice command (confirmation) then release the object
DetectPersonHIdentify a person from camera images
WalkingSupportHMove toward the user, rotate until user can grasp the handle, then drive the robot according to the user movements
BringObjectHDetectKnownObject, GraspKnownObject, PutObjectOnTray, DriveToUser
CleanTableHHandoverObject detect all objects on a table, put onto tray, carry to kitchen, then put into kitchen sink
CleanWindowHPut sponge/brush onto tray, drive toward window, perform cleaning motions
Tab.1  Example skills defi ned for the domestic robot.
Fig.7  Example sequence diagram for the bring-object service. Following the user request (“bring me milk”), the planner first checks the CAM module to find the position of the selected object, and then plans and calls the corresponding services on the domestic robot. In this case, the two drive commands (first to the refrigerator, and then back to the user) are handled by the MIRA/CogniDrive module, while perception and arm motions are handled by ROS.
Fig.8  Overview of the multi-modal perception pipeline. The data from the cameras and the Xtion RGB-D sensor is forwarded to the SIFT-based recognition of known objects and a marker-based pose-recognition algorithm. Depth images and point-cloud data from the Xtion sensor are processed using the tabletop_object_recognition stack to find clusters that correspond to (graspable) objects. The recognition results of the different pipelines are then fed into the central Visionhub node, which performs data fusion and outputs a list of detected objects, together with the object pose information and covariance estimates. Point-cloud data is also forwarded to the person-detection modules and CogniDrive for 3D-obstacle avoidance while driving.
Fig.9  Object perception in a cluttered environment. (a) Original robot camera image showing several boxes on a kitchen table. Note that the objects overlap and are only partially visible. (b) Visualization of the robot together with the recognized graspable objects in the ROS RViz tool. For every detected object, the object name, shape, and the full 6-DOF pose (position and orientation) of the object center are calculated and published on ROS messages. In the example scene, all four boxes have been detected correctly by the Visionhub node, and the object shapes (cyan boxes) and object poses (red, green, and blue axes) are visualized.
Fig.10  Object exchange between the domestic and the condominium robots. (a) AprilTag markers are used on the box; (b) the round handle inside the box improves grasp stability for the Jaco hand.
Fig.11  Photos from the first experimental loop. (a) Tablet-based user interface; (b) condominium robot; (c) indoor walking support; (d) garbage exchange between domestic and condominium robots.
Fig.12  Typical results from the first large-scale experimental test of the Robot-Era system. The usability and acceptability of the three robots and the services were analyzed based on questionnaires (using the SUS score) and video analysis of the experiments. Overall scores are promising, despite bugs and problems during certain runs of the experiments. Scores of 65 and 85 points were used to split the classes. (a) Acceptance of DORO appearance; (b) usability of “shop and drug delivery service”; (c) acceptance of “shop and drug delivery service”; (d) usability of garbage collection; (e) acceptance of garbage collection; (f) usability of indoor walking support.
1 I. Conrad, S. Glende, L. Krezdorn, L. Trieste, G. Turchetti. D9.4, Second report on services, business and marketing models for Robot-Era services, project report, 2015, www.robot-era.eu
2 F. Cavallo, M. Aquilano, M. C. Carrozza, P. Dario. Robot-era project: The vision of 3D service robotics. Gerontechnology, 2012, 11(2): 364
3 F. Marcellini,  Report on the robotic services analysis with respect to elderly users, project report, 2012, www.robot-era.eu
4 N. J. Nilsson. Shakey the Robot (Technical Note 323). Menlo Park, California: SRI International, Artificial Intelligence Center, 1984
5 R. Brooks. A robust layered control system for a mobile robot. IEEE J. Robot. Autom., 1986, 2(1): 14–23
6 T. H. J. Collett, B. A. MacDonald, B. P. Gerkey. Player 2.0: Toward a practical robot programming framework. In: Proceedings of Australasian Conference on Robotics and Automation (ACRA 05). IEEE, 2005: 1–9
7 G. Metta, P. Fitzpatrick, L. Natale. YARP: Yet another robot platform. Int. J. Adv. Robot. Syst., 2006, 3(1): 43–48
8 M. Quigley,  ROS: An open-source Robot Operating System. In: International Conference on Robotics and Automation (ICRA) Workshop on Open-Source Software, 2009
9 E. Einhorn, T. Langner, R. Stricker, C. Martin, H. M. Gross. MIRA-middleware for robotic applications. In: Proceedings of 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2012), 2012: 2591–2598
10 R. D. Schraft, J. Neugebauer, C. Schaeffer, T. May. Care-O-bot®: Ein technisches Hilfssystem für unterstützungs- und pflegebedürftige Personen im häuslichen Bereich. In: H. Wörn, R. Dillmann, D. Henrich, eds. Autonome Mobile Systeme 14. Berlin, Heidelberg: Springer, 1998: 234–244
11 M. Hans, B. Graf. Robotic home assistant Care-O-bot II. In: E. Prassler, , eds. Advances in Human-Robot Interaction. Berlin, Heidelberg: Springer, 2004: 371–384
12 C. Parlitz, M. Hägele, P. Klein, J. Seifert, K. Dautenhahn. Caro-O-bot 3—Rationale for human-robot interaction design. In: Proceedings of 39th International Symposium on Robotics (ISR), 2008: 275–280
13 M. Hans, W. Baum. Concept of a hybrid architecture for Care-O-bot. In: Proceedings of 10th IEEE International Workshop on Robot and Human Interactive Communication. IEEE, 2001: 407–411
14 T. Morita, H. Iwata, S. Sugano. Development of human symbiotic robot: WENDY. In: Proceedings of 1999 IEEE International Conference on Robotics and Automation, vol. 4. IEEE, 1999: 3183–3188
15 Willow Garage Inc. Robot for research and innovation. https://www.willowgarage.com/pages/pr2/overview
16 S. Cousins. ROS on the PR2 [ROS Topics]. Robot. Autom. Mag., IEEE, 2010, 17(3): 23–25
17 A. Hermann,  Hardware and software architecture of the bimanual mobile manipulation robot HoLLiE and its actuated upper body. In: Proceedings of 2013 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM). IEEE, 2013: 286–292
18 MetraLabs GmbH. Scitos G5 mobile platform. metralabs.com/index.php?option=com_content&view=article&id=70&Itemid=64
19 F. Cavallo. Improving domiciliary robotic services by integrating the ASTRO robot in an AmI infrastructure. In: F. Röhrbein, G. Veiga, C. Natale, eds. Gearing Up and Accelerating Cross—Fertilization between Academic and Industrial Robotics Research in Europe. Springer, 2014: 267–282
20 Schunk Gmbh & Co. KG. Powerball lightweight arm LWA 4P. http://mobile.schunk-microsite.com/en/produkte/produkte/powerball-lightweight-arm-lwa-4p.html
21 Kinova Robotics, Kinova Jaco Research Edition. kinovarobotics.com/products/jaco-researchedition/
22 Schunk Gmbh & Co. KG. Dextrous lightweight arm LWA 4D. http://mobile.schunk-microsite.com/en/produkte/produkte/dextrous-lightweight-arm-lwa-4d.html
23 BionicRobotics GmbH. Bionic robotics: Biorob arm. http://www.bionic-robotics.de/en/products/biorob-arm.html
24 Universal Robots. UR5 arm. www.universal-robots.com/GB/Products/
25 V. Maheu, J. Frappier, P. S. Archambault, F. Routhier. Evaluation of the JACO robotic arm: Clinico-economic study for powered wheelchair users with upper-extremity disabilities. In: Proceedings of 2011 IEEE International Conference on Rehabilitation Robotics (ICORR). IEEE, 2011: 1–5
26 A. Saffiotti, M. Broxvall. PEIS ecologies: Ambient intelligence meets autonomous robotics. In: Proceedings of the International Conference on Smart Objects and Ambient Intelligence, 2005: 277–281
27 M. Broxvall, M. Gritti, A. Saffiotti, B. S. Beo, Y. J. Cho. PEIS ecology: Integrating robots into smart environments. In: Proceedings of 2006 IEEE International Conference on Robotics and Automation (ICRA), 2006: 212–218
28 M. Di Rocco,  A planner for ambient assisted living: From high-level reasoning to low-level robot execution and back. In: AAAI Spring Symposium “Qualitative Representations for Robots”, 2014: 10–17
29 N. Hendrich, H. Bistry, B. Adler, J. Zhang. User-driven software design for an elderly care service robot. In: Proceedings of the 8th International Conference on Pervasive Computing Technologies for Healthcare, 2014: 142–149
30 I. Sucan. Universal robot description format. http://wiki.ros.org/urdf
31 E. Marder-Eppstein, V. Pradeep. ROS actionlib. http://wiki.ros.org/actionlib
32 D. G. Lowe. Object recognition from local scale-invariant features. In: The Proceedings of the Seventh IEEE International Conference on Computer Vision, vol. 2. IEEE, 1999: 1150–1157
33 E. Olson. AprilTag: A robust and flexible visual fiducial system. In: Proceedings of 2011 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2011: 3400–3407
34 J. R. Lewis, J. Sauro. The factor structure of the system usability scale. In: M. Kurosu, ed. Human Centered Design. Berlin, Heidelberg: Springer, 2009: 94–103
35 S. McLellan, A. Muddimer, S. C. Peres. The effect of experience on System Usability Scale ratings. J. Usability Stud., 2012, 7(2): 56–67
36 F. Cavallo, D8.2 Report on results obtained in the first cycle of the Robot-Era experimentation, project report, 2014. www.robot-era.eu
37 F. Pecora,  D2.8 Second report on the S/T requirements of the Robot-Era components, project report, 2015. www.robot-era.eu
[1] Ji Zhou, Jianfeng Chen. Aiming to Be a World-Class Journal Leading in Innovation: A Letter from the Editors-in-Chief[J]. Engineering, 2018, 4(5): 581-.
[2] Suping Peng, Jianghai Xia, Jiulong Cheng. Applications of Geophysics in Resource Detection and Environmental Protection[J]. Engineering, 2018, 4(5): 584-585.
[3] Takashi Asaeda, Senavirathna Mudalige Don Hiranya Jayasanka, Li-Ping Xia, Abner Barnuevo. Application of Hydrogen Peroxide as an Environmental Stress Indicator for Vegetation Management[J]. Engineering, 2018, 4(5): 610-616.
[4] Zhifeng Yang. Watershed Ecology and Its Applications[J]. Engineering, 2018, 4(5): 582-583.
[5] Neil Andrew. Engineers: The Key to Sustainable Water Use[J]. Engineering, 2018, 4(5): 586-587.
[6] Jianbo Shen, Fusuo Zhang, Kadambot H.M. Siddique. Sustainable Resource Use in Enhancing Agricultural Development in China[J]. Engineering, 2018, 4(5): 588-589.
[7] Yurui Fan, Guohe Huang, Yin Zhang, Yongping Li. Uncertainty Quantification for Multivariate Eco-Hydrological Risk in the Xiangxi River within the Three Gorges Reservoir Area in China[J]. Engineering, 2018, 4(5): 617-626.
[8] Nikita Shivhare, Prabhat Kumar Singh Dikshit, Shyam Bihari Dwivedi. A Comparison of SWAT Model Calibration Techniques for Hydrological Modeling in the Ganga River Watershed[J]. Engineering, 2018, 4(5): 643-652.
[9] Jeffrey G. Paine, Edward W. Collins, Lucie Costard. Spatial Discrimination of Complex, Low-Relief Quaternary Siliciclastic Strata Using Airborne Lidar and Near-Surface Geophysics: An Example from the Texas Coastal Plain, USA[J]. Engineering, 2018, 4(5): 676-684.
[10] John H. Bradford, Janna Privette, David Wilkins, Richard Ford. Reverse-Time Migration from Rugged Topography to Image Ground-Penetrating Radar Data in Complex Environments[J]. Engineering, 2018, 4(5): 661-666.
[11] Jishan He. Combined Application of Wide-Field Electromagnetic Method and Flow Field Fitting Method for High-Resolution Exploration: A Case Study of the Anjialing No. 1 Coal Mine[J]. Engineering, 2018, 4(5): 667-675.
[12] Edward W. Woolery. SH-Mode Seismic-Reflection Imaging of Earthfill Dams[J]. Engineering, 2018, 4(5): 694-701.
[13] Massimiliano Lega, Marco Casazza, Laura Turconi, Fabio Luino, Domenico Tropeano, Gabriele Savio, Sergio Ulgiati, Theodore Endreny. Environmental Data Acquisition, Elaboration and Integration: Preliminary Application to a Vulnerable Mountain Landscape and Village (Novalesa, NW Italy)[J]. Engineering, 2018, 4(5): 635-642.
[14] Jiulong Cheng, Guangdong Song, Xiaoyun Sun, Laifu Wen, Fei Li. Research Developments and Prospects on Microseismic Source Location in Mines[J]. Engineering, 2018, 4(5): 653-660.
[15] Feng Cheng, Jianghai Xia, Chao Shen, Yue Hu, Zongbo Xu, Binbin Mi. Imposing Active Sources during High-Frequency Passive Surface-Wave Measurement[J]. Engineering, 2018, 4(5): 685-693.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
国内刊号:CN10-1244/N    国际刊号:ISSN2095-8099
版权所有 © 2015 高等教育出版社  《中国工程科学》杂志社
京ICP备11030251号-2

 Engineering