First steps of robotic perception: the turning point of the 1990s

Proceedings of the IEEE - Tập 90 Số 7 - Trang 1094-1112 - 2002
B.Y. Zavidovique1
1Institut dE28099Electronique Fondamentale, Paris XI University, Orsay, France

Tóm tắt

In this paper we analyze the early evolution of robot perception toward robot autonomy: respective impacts of the bare technology and of advanced control are put in perspective. At first, the UniBuM's vehicle, operational in the 1990s, is taken as a study example. The strong points of such systems suggest a discussion of the software technology in image processing and of the hardware technological concept of silicon retina regarding their respective contribution to real machine vision. This second part elicits limitations, which lead again to concepts and realizations in control, through the need for sensor fusion and active vision. The ultimate step to robot autonomy would be learning, and this is considered in the conclusion.

Từ khóa

#Robots #Turning #Humans #Vehicles #Silicon #Sensor fusion #Layout #Data acquisition #Data mining #Cameras

Tài liệu tham khảo

10.1007/978-1-4615-5965-8_12 10.1109/34.211463 10.1109/5.5968 10.1007/BF01418982 10.1109/ICPR.1996.546162 10.1016/0004-3702(91)90080-4 10.1109/CVPR.1994.323849 10.1007/978-1-4899-5379-7 noton, 1971, eye movement and visual perception, Sci Amer, 224, 34 10.1109/34.494639 merlo, 1987, optimal control of a robot perception system, Proc IASTED Int Symp GENEVE lacroix, 1993, perception for autonomous navigation in a natural environment, Proc Workshop on Computer Vision for Space Applications moutarlier, 1989, multisensory data fusion for mobile robot location and 3d modeling, Proc IARP 1st Workshop on Multi-Sensor Fusion and Environment Modeling 10.1109/ROBOT.1993.291956 10.1109/ROBOT.1992.220337 hotz, 1993, incremental construction of local dem for an autonomous planetary rover, Proc INRIA-ESA Workshop on Computer Vision for Space Applications, 33 devy, 1991, environment modeling from a laser/camera multisensor system, Proc IARP 2nd Workshop on Multi-Sensor Fusion and Environment Modeling moutarlier, 1989, stochastic multisensory data fusion for mobile robot location and environment modeling, Proc Int Symp Robotics Research moutarlier, 1989, an experimental system for incremental environment modeling by an autonomous mobile robot, Proc Int Symp Experimental Robotics 10.1109/IROS.1993.583192 nashashibi, 1994, 3d autonomous navigation in a natural environment, Proc IEEE Int Conf Robotics and Automation tsosos, 1987, a complexity level analysis of vision, Proc IEEE Int Conf Computer Vision ICCV 87 10.1007/BF01539627 10.1006/cviu.1994.1039 10.1006/ciun.1994.1031 10.1109/ICPR.1990.118128 10.1007/BF00058749 10.1016/0262-8856(90)90050-F 10.1109/ROBOT.1992.220292 10.1109/ICCV.1990.139576 10.1006/cviu.1994.1045 fillatreau, 1993, modeling of unstructured terrain and feature extraction using b-spline surfaces, Proc 93 Int Conf Advanced Robotics (ICAR) fillatreau, 1993, localization of an autonomous mobile robot from 3d depth images using heterogeneous feature, Proc IEEE International Workshop on Intelligent Robots and Systems 10.1142/S0218001493000042 grandjean, 1989, 3-d modeling of indoor scenes by fusion of noisy range and stereo data, Proc IEEE Int Conf Robotics and Automation, 681 10.1109/IROS.1993.583117 bauzil, 1992, Robot mobile HILARE II Mat&#x00E9 riel et biblioth&#x00E8 ques associ&#x00E9 es bauzil, 1991, Robot mobile HILARE II Description du mat&#x00E9 riel boissier, 1991, autonomous planetary rover (vap): the robotics concepts, Proc IARP 91 Robotics in Space 10.1109/IROS.1992.594565 devy, 1992, localization of an autonomous mobile robot from 3d depth images in a structured environment, Proc 23rd Int Symp on Industrial Robots (ISIR 92), 407 chatila, 1984, L exp&#x00E9 rience du robot mobile HILARE xqxqxq, 1988, xqxqxq, Proc DARPA Workshop on Image Understanding Benchmark duff, 1981, Languages and Architectures for Image Processing 10.1016/0010-0285(80)90005-5 olivier, 1993, Strat&#x00E9 gies d acquisition de traitement et de prise en compte d informations pour contr&#x00F4 1e de robot en environnement non structur&#x00E9 crisman, 1992, color region tracking for vehicle guidance, Active Vision, 107 10.1016/0010-0277(84)90023-4 10.1109/MAP.1996.491294 dessoude, 1993, Contr&#x00F4 le Perceptif en milieu hostile Allocation de ressources automatique pour un syst&#x00E8 me multicapteur zavidovique, 1990, smart retina, Proc ICOGNITIVA 90, 495 harris, 1989, an analog vlsi chip for thin-plate surface interpolation, Neural Information and Processing Systems, 687 10.1109/ICPR.1988.28383 zavidovique, 1981, bilevel processing of multilevel pictures, Proc Pattern Recognition and Image Processing Conf birminwal, 1984, dual control guidance for simultaneous identification and interception, Automatica, 20 10.1016/S0146-664X(78)80027-6 mohr, 1986, outils logiciels pour le traitement des images, Proc 2nd CESTA, 201 chassery, 1986, ips design and abilities, Proc 8th ICPR, 913 lay, 1986, morpholog: an image processing software package, Proc CAPAIDM, 463 tamura, 1983, design and implementation of spider: a transportable i.p. software package, Proc CVGIP, 23 3, 273 10.1016/B978-0-444-86325-6.50008-4 duff, 1986, Intermediate-Level Image Processing duff, 1983, Computing Structures for Image Processing ballard, 1982, Computer Vision boukir, 1993, "Reconstruction 3D d un Environnement Statique par Vision Active " 10.1016/0262-8856(95)98754-H 10.1109/5.5964 merlo, 1988, Techniques probabilistes d int&#x00E9 gration et de contr&#x00F4 1e de la perception en vue de son exploitation par le syst&#x00E8 me de d&#x00E9 cision d un robot bajcsy, 1985, active perception vs. passive perception, IEEE Proceedings, 55 10.1007/BF00133571 anderson, 1979, Optimal Filtering coutelle, 1995, Conception d un syst&#x00E8 me &#x00E0 base d op&#x00E9 rateurs de vision rapides lindberg, 1976, Theories of Vision from Al-Kindi to Kepler 10.1115/1.3662552 10.1109/34.121789 dickmans, 1989, simulation for the development of a visual autopilot-system for road vehicles, Automotive Simulation, 11 10.1016/0952-1976(91)90044-7 10.1007/3-540-57233-3_1 zavidovique, 1987, robot perception systems: some design issues, NATO Adv Res Workshop on Real Time Object and Environment Measurement &amp Classifications, 93 10.1007/BF01212361 jochem, 1993, maniac: a next generation neurally based autonomous road follower, Proc Int Conf Intelligent Autonomous Systems IAS-3 10.1007/978-3-642-83325-0_6 dickmanns, 1992, expectation-based dynamic scene understanding, Active Vision, 303 10.1109/TPAMI.1986.4767851 10.1016/0146-664X(75)90012-X kories, 1984, a class of stable feature extractors for time-varying imagery, Proc 7th ICPR, 919 moravec, 1977, toward automatic visual obstacle avoidance, 5th IJCAI haralick, 1984, survey: image segmentation techniques, Proc CVGIP, 29, 100 10.1016/0031-3203(81)90028-5 pavlidis, 1986, a critical survey of image analysis methods, Proc 8th ICPR, 502 marr, 1982, Vision 10.1109/ICPR.1990.119441 eckstein, 1986, psiwag: a language for logic programming in image analysis, Proc 8th ICPR, 1117 clermont, 1990, Parallel programming of a pyramid machine Application to image segmentation 10.1016/0146-664X(79)90081-9 crookes, 1989, an algebra-based language for image processing on transputers, Image Processing and its Applications 1989 Third International Conference on, 457 10.1109/CVPR.1988.196318 matheron, 1975, Random Sets and Integral Geometry serra, 1982, Image Analysis and Mathematical Morphology ritter, 1987, image algebra in a “nutshell, Proc 1st ICCV chassery, 1990, Images m&#x00E9 thodologie et applications krusemark, 1983, an operating system interface for transportable ip software & image random file access routines, Proc CVGIP, 23 4, 6–66 10.1016/0167-8655(85)90018-2 serfaty, 1985, programming facilities in image processing, Proc Compint, 804 dacic, 1990, Conception et exploitation intuitive de syst&#x00E8 mes informatiques complexes borgna, 1983, VISIR un syst&#x00E8 me pour la VISIon en Robotique g&#x00E9 n&#x00E9 rale draman, 1986, icotech: poste de traitement d'images configurable, Proc 2nd CESTA, 437 tomita, 1979, extraction of multiple regions by smoothing in selected neigborhoods, Proc SMC, 7, 107 10.1016/0146-664X(79)90102-3 10.1007/978-3-642-82940-6 fernandez, 1990, languagues for a line processor: sympati2, Proc COGNITIVA, 485 10.1109/VL.1994.365602 10.1109/56.782 10.1016/0743-7315(92)90104-U 10.1109/CVPR.1992.223232 10.1007/3-540-55426-2_63 10.1007/BF01421204 10.1109/ICCV.1993.378188 pesty, 1988, kids: knowledge-based image diagnosis system, Proc 2nd TIPI, xxxl, 1 10.1109/ICPR.1988.28187 10.1109/TDSCEN.1989.68113 hasegawa, 1986, automated construction of image processing procedures by sample-figure presentation, Proc 6th lCPR, 586 10.1109/ICCV.1993.378189 serfaty, 1991, Les sources de connaissances en Vision le point de vue Traitement d Images 10.1109/CCV.1988.590032 taniguchi, 1989, knowledge based image processing system: ipssens-ii, Proc 3rd IPA 10.1109/TPAMI.1984.4767570 amat, 1986, design of a vision system in remote sensing: modeling image processing sequences, Proc 8th IJCPR, 1244 sakuae, 1985, automatic generation of image processing programs by knowledge based verification, IEEE Proc CVPR, 189 thonnat, 1985, design of an expert system for object classification through an application to the classification of galaxies, IEEE Proc CVPR, 206 yamamoto, 1990, miracle-iv: multiple image recognition system toward concept learning-intelligent vision, Proc 9th JCPR, 818 devy, 1986, visir: un système de développement spécialisé pour le traitement d'images en robotique avancée, Proc 2nd CESTA, 374 10.1016/0734-189X(84)90199-3 10.1109/ICPR.1990.119431 frécaut freacaut, 1987, Int&#x00E9 gration du concept d aide &#x00E0 l utilisateur dans les syst&#x00E8 mes logiciels complexes jeansoulin, 1985, inrimagesm: logiciel de traitement d'image sous unix, Proc &#x201C Journ&#x00E9 es SM90&#x201D, 352 10.1109/ICPR.1990.119440 10.1109/ICPR.1990.119335 chassery, 1986, expert systems, image processing and image interpretation, Proc 8th ICPR, 175 matsuyama, 1990, SJGMA A Knowledge-Based Aerial Image Understanding System garbay, 1986, knowledge and strategies for image segmentation, 8th ICPR, 669 10.1109/ICCV.1993.378187 10.1016/0167-8655(90)90114-H 10.1108/eb007671 10.1109/ICCV.1993.378236 10.1109/ICCV.1993.378186 10.1109/CVPR.1992.223233 10.1016/1049-9660(92)90084-G 10.1016/0262-8856(92)90010-Z wixson, 1992, Exploiting World Structure to Efficiently Search for Objects wixson, 1994, Gaze Selection for Visual Search kutulakos, 1992, recovering shape by purposive viewpoint adjustment, Int J Comput Vis, 12, 113, 10.1007/BF01421200 pissard-gibollet, 1991, Asservissement visuel appliqu&#x00E9 &#x00E0 un robot mobile Etat de 1 art et mod&#x00E9 lization cin&#x00E9 matique rimey, 1993, Control of selective perception using Bayes nets and decision theory rimey, 1993, Goal-oriented dynamic vision isard, 1996, contour tracking by stochastic propagation of conditional density, Proc Eur Conf Computer Vision, 343 isard, 1998, condensation: unifying low-level and high-level tracking in a stochastic framework, Proc 5th Eur Conf Computer Vision, 10.1023/A:1008078328650 popper, 1972, Objective Knowledge An Evolutionary Approach 10.1364/AO.28.003810 wolffenbuttel, 1987, photodiodes in silicon with an intrinsic color filtering capability, Proc 4th Int Conf Solid State Sensors and Actuators, 219 kataoka, 1986, three-dimensional integrated sensors, 1986 International Electron Devices Meeting, 361, 10.1109/IEDM.1986.191192 kioi, 0, monolithic character recognition system implemented as prototype intelligent image sensor by 3d integration technology, Proc IEDM88 Tech Dig, 66 10.1109/ISCAS.1988.15146 mead, 0, a sensitive electronic photoreceptor, Proc 1985 Chapel Hill Conf VLSI, 463 10.1109/31.16562 hu, 1988, eeproms as analog storage devices for neural nets, Neural Networks 1 mead, 1989, adaptive retina, Analog VLSI Implementation of Neural Systems, 10.1007/978-1-4613-1639-8_10 10.1109/ICASSP.1990.116041 harris, 1999, resistive fuses: analog hardware for detecting discontinuities in early vision, Analog VLSI Implementation of Neural Systems, 27 10.1109/CICC.1991.164120 glasser, 0, a uv write-enabled prom, Proc 1985 Chapel Hill Conf VLSI, 61 10.1109/72.80329 mahowald, 1991, silicon retina with adaptive photo-receptors, Visual Information Processing From Neurons to Chips, 10.1117/12.45540 fossum, 1984, a linear and compact charge-coupled charge packet differencer/replicator, IEEE Transactions on Electron Devices, 31, 1784, 10.1109/T-ED.1984.21789 fossum, 1987, charge-coupled computing for focal plane image preprocessing, J Opt Eng, 26, 916, 10.1117/12.7974171 10.1109/JSSC.1968.1049925 sage, 1985, a high speed analog two-dimensional gaussian image convolver, Proc OSA Topical Meeting Machine Vision 10.1109/ICASSP.1991.150985 koch, 86, analog “neuronal” network in early vision, Proc Nat Acad Sci USA83, 4263, 10.1073/pnas.83.12.4263 moravec, 2000, Robots Re-Evolving Mind 10.1162/neco.1989.1.2.184 moravec, 1998, when will computer hardware match the human brain?, J Transhumanism, 1 10.1007/978-3-642-68402-9_1 10.1007/978-1-4899-1004-2_2 offen, 1985, VLSI Image Processing 10.1016/0167-6911(86)90048-4 10.1109/52.103576 10.1002/ima.1850020304 10.1109/ICPR.1990.119434 10.1109/2.30729 tanner, 1986, an integrated optical motion sensor, VLSI Signal Processing, 59 mead, 1988, Analog VLSI and Neural Systems tanner, 1984, a correlating optical motion detector, Proc Conf Advanced Res VLSI, 57 10.1109/21.61200 hebert, 1989, terrain mapping for a roving planetary explorer, Proc IEEE Int Conf Robotics and Automation, 997 10.1109/CVPR.1992.223164 10.1109/34.121795 kweon, 1991, Modeling rugged terrain by mobile robots with multiple sensors 10.1109/ROBOT.1988.12144 brooks, 1979, the acronym model-based vision system, Proc 6th Int Joint Conf Artificial Intelligence, 105 beaudet, 1989, time and space multiplexing focal plane convolvers, Proc SPIE, 1071, 10.1117/12.952510 10.1109/4.104177 10.1016/0004-3702(81)90028-X binford, 1971, visual perception by a computer, Proc IEEE Conf Systems and Control delbrück delbruck, 1988, an electronic photoreceptor sensitive to small changes in intensity, Advances in Neural Information Processing Systems I, 720 brooks, 1980, representing and reasoning about specified scenes, Proc DARPA IU Workshop, 95 10.1109/WVM.1991.212769 10.1109/ISCAS.1991.176627 umminger, 1989, implementing gradient following in analog vlsi, Proc Decennial Caltech Conf on VLSI, 195 deweerth, 1988, a two-dimensional visual tracking array, Proc MIT Conf VLSI, 259 10.1109/ROBOT.1991.131797 10.1109/2.31 10.1117/12.968987 carson, 1988, applications of advanced “z” technology focal plane architecture, Proc SPIE, 930, 164, 10.1117/12.946636