Robot Vision SS 2005 Matthias Rüther 2 Organization VO: Tuesday 14:15-15:45 Seminarraum ICG Exam: Written Exam Oral Exam if Requested KU:implementation of lecture topics in the real world (on the lab-robots) Groups of three students Possible problems on the last slide Scheduling of topics: 8.3.2005 If you are interested: excursions to industrial vision companies (Alicona Imaging, M&R)
Robot Vision SS 2005 Matthias Rüther 3 Time Table 1.3. : Introduction and Overview 8.3. : Projective Geometry (1) 15.3. : Projective Geometry (2) 12.4. : Projective Geometry (3) 19.4. : Projective Geometry (4) 26.4. : Camera Technologies 3.5. :Shape From X (1) 10.5. : Shape From X (2) 24.5. : Shape From X (3) 31.5. : Robot Kinematics (1) 7.6. : Robot Kinematics (2) 14.6. : Tracking of Moving Objects 21.6. : Visual Servoing / Hand Eye Coordination
Robot Vision SS 2005 Matthias Rüther 4 Literature Sciavicco, L., Siciliano, B., Modelling and Control of Robot Manipulators 2nd Ed., Springer, 2000 Ballard D.H., Brown C.M., "Computer Vision", Prentice-Hall, 1982 Sonka M., Hlavac V., Boyle Image Processing, Analysis and Machine Vision, Chapman Hall, 1998 Nalva V.S., "A Guided Tour of Computer Vision", Addison-Wesley Publishing Company, 1993 Horn B.K.P., "Robot Vision", MIT Press, Cambridge, 1986 Shirai Y., "Three- Dimensional Computer Vision", Springer Verlag, 1987 Faugeras O., Three-Dimensional Computer Vision A Geometric Viewpoint, MIT Press, 1993 Hartley R., Zissermann A., Multiple View Geometry in Computer Vision, Cambridge, 2001.
Robot Vision SS 2005 Matthias Rüther 5 Robotics What is a robot? "A reprogrammable, multifunctional manipulator designed to move material, parts, tools, or specialized devices through various programmed motions for the performance of a variety of tasks" Robot Institute of America, 1979 … in a three-dimensional environment. Industrial –Mostly automatic manipulation of rigid parts with well-known shape in a specially prepared environment. Medical –Mostly semi-automatic manipulation of deformable objects in a naturally created, space limited environment. Field Robotics –Autonomous control and navigation of a mobile vehicle in an arbitrary environment.
Robot Vision SS 2005 Matthias Rüther 6 Experimental/Industrial/Commercial Robots
Robot Vision SS 2005 Matthias Rüther 7 Industrial Robots
Robot Vision SS 2005 Matthias Rüther 8 Challenging Environments
Robot Vision SS 2005 Matthias Rüther 9 Service and Assistance
Robot Vision SS 2005 Matthias Rüther 10 FRIEND Project
Robot Vision SS 2005 Matthias Rüther 11 Robot vs Human Robot Advantages : –Strength –Accuracy –Speed –Does not tire –Does repetitive tasks –Can Measure Human advantages: –Intelligence –Flexibility –Adaptability –Skill –Can Learn –Can Estimate
Robot Vision SS 2005 Matthias Rüther 12 Robotics: Goals and Applications Robotics does not intend to develop the artificial human! [ Whitney, D. E., Lozinski, C. A. and Rourke, J. M. (1986) Industrial robot forward calibration method and results. ] Goal: combine robot and human abilities. Applications: –Automation (Production) –Inspection (Quality control) –Remote Sensing (Mapping) –Man-Machine interaction („Cobot“) –Robot Companion (Physically challenged people) –See [Brady, M. et. al. (eds). „Robot Motion: Planning and Control“]
Robot Vision SS 2005 Matthias Rüther 13 What can Computer Vision do for Robotics? Accurate Robot-Object Positioning Keeping Relative Position under Movement Visualization / Teaching / Telerobotics Performing measurements Object Recognition (see LV „Bildverarbeitung u. Mustererkennung“, „Bildverstehen“, „AK Computer Vision“) Registration Visual Servoing
Robot Vision SS 2005 Matthias Rüther 14 Combining Computer Vision and Robotics Abstraction level Motor Modeling : what voltage should I set now ? Control (PID) : what voltage should I set over time ? Kinematics : if I move this motor somehow, what happens in other coordinate systems ? Motion Planning : Given a known world and a cooperative mechanism, how do I get there from here ? Bug Algorithms : Given an unknowable world but a known goal and local sensing, how can I get there from here? Mapping : Given sensors, how do I create a useful map? Localization : Given sensors and a map, where am I ? low high Vision : If my sensors are eyes, what do I do?
Robot Vision SS 2005 Matthias Rüther 15 Computer Vision What is Computer Vision? "Computer Vision describes the automatic deduction of the structure and the properties of a (possible dynamic) three- dimensional world from either a single or multiple two-dimensional images of the world" [Nalva VS, "A Guided Tour of Computer Vision"] Measurement –Measure shape and material properties in a 3D environment. Accuracy is important. Recognition –Cognitive systems interpret a 3D environment (object classification, categorization). Systems are allowed to fail to a certain extent (similar to humans). Navigation –Navigation Systems orient themselves in a 3D environment. Robustness and time are important.
Robot Vision SS 2005 Matthias Rüther 16 Measurement „Shape from X“ techniques measure shape properties of objects from 2D digital images. –Shape from Stereo: two cameras obeserve an object from different viewpoints (similar to human eye). –Shape from focus: limited depth of focus allows to measure object- camera-distance. –Shape from structured light: a light pattern is projected on the object, the pattern deformation gives shape information. –Shape from Shading: an object is illuminated from a single direction. Light reflection depends on object shape and follows a reflectance function.
Robot Vision SS 2005 Matthias Rüther 17 Shape from Stereo
Robot Vision SS 2005 Matthias Rüther 18 Shape from Stereo
Robot Vision SS 2005 Matthias Rüther 19 Shape from Focus
Robot Vision SS 2005 Matthias Rüther 20 Shape from Structured Light Structured Light Sensor Figures from PRIP, TU Vienna
Robot Vision SS 2005 Matthias Rüther 21 Shape from Shading
Robot Vision SS 2005 Matthias Rüther 22 Navigation SLAM: Simultaneous Localization and Mapping. –Where am I on my map? –If the place is unknown, build a new map, try to merge it with the original map. Visual Odometry: calculate the relative motion of the camera between two frames. Summing up the motion gives the camera path. Error propagation! Visual Servoing: move to / maintain a relative position between robot end effector and an object. Tracking: continuously measure the position of an object within the sensor coordinate frame.
Robot Vision SS 2005 Matthias Rüther 23 SLAM Mapping:
Robot Vision SS 2005 Matthias Rüther 24 SLAM The final map:
Robot Vision SS 2005 Matthias Rüther 25 SLAM Navigation:
Robot Vision SS 2005 Matthias Rüther 26 Visual Odometry
Robot Vision SS 2005 Matthias Rüther 27 Visual Servoing
Robot Vision SS 2005 Matthias Rüther 28 Tracking
Robot Vision SS 2005 Matthias Rüther 29 Tracking
Robot Vision SS 2005 Matthias Rüther 30 Registration Registration of CAD models to scene features: Figures from P.Wunsch: Registration of CAD-Models to Images by Iterative Inverse Perspective Matching
Robot Vision SS 2005 Matthias Rüther 31 KU: Student Problems Shape from Stereo3 students Shape from Focus3 students Shape from Structured Light:Laser3 students Shape from Structured Light:Pattern3 students Shape from Shading3 students Robot Kinematics3 students 2D Grip Planning2..3 students 2D Visual Servoing3 students 2D Tracking3 students Registration / Model Fitting3 students Visual Odometry + Randomized RANSAC3 students