Abstract
This paper addresses the problem of autonomous quadrotor navigation through a previously-mapped indoor area. In particular, we focus on the case where a user walks through a building and collects images. Subsequently, a visual map of the area, represented as a graph of linked images, is constructed and used for automatically determining visual paths (i.e., sequences of images connecting the start to the end image locations specified by the user). The quadrotor follows the desired path by iteratively (i) determining the desired motion to the next reference frame, (ii) controlling its roll, pitch, yaw-rate, and thrust, and (iii) appropriately switching to a new reference image. For motion estimation and reference-image switching, we concurrently employ the results of the 2pt and the 5pt RANSAC to distinguish and deal with both cases of sufficient and insufficient baseline (e.g., rotation in place). The accuracy and robustness of our algorithm are evaluated experimentally on two quadrotors navigating along lengthy corridors, and through tight spaces inside a building and in the presence of dynamic obstacles (e.g., people walking).
This work was supported by the AFOSR (FA 9550-10-1-0567).
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
The 2pt RANSAC estimates the relative orientation \(^{\scriptscriptstyle {I_{1}}}_{\scriptscriptstyle {I_{2}}}\mathbf {R}\) between two images, \(I_{\scriptscriptstyle {1}}\) and \(I_{\scriptscriptstyle {2}}\), under the assumption of very small baseline compared to the depth of the scene. A closed-form solution for the 2pt minimal case is provided in Sect. 6.2, while the analytical solution for the least-squares solver is presented in [21]. The 5pt RANSAC [26] estimates the relative orientation \(^{\scriptscriptstyle {I_{1}}}_{\scriptscriptstyle {I_{2}}}\mathbf {R}\) and the unit vector of translation \(^{\scriptscriptstyle {I_{1}}}\mathbf {t}_{\scriptscriptstyle {I_{2}}}\) between two images \(I_{\scriptscriptstyle {1}}\) and \(I_{\scriptscriptstyle {2}}\).
- 2.
Under low-light conditions, the velocity measurements are reliable only for a fixed tilt angle of the vehicle. Note that when in motion, the quadrotor changes its roll and pitch which causes image blurriness (due to the increased exposure) and, hence, large errors in the optical-flow estimates.
- 3.
Note that although both the embedded controller and the cell phone contain IMUs, which can be used, in conjunction with the camera, to form a vision-aided inertial navigation system [18], in this work, we intentionally focus on a “light”, in terms of processing, vision-only approach so as to assess its performance and use it as a baseline for future comparisons.
- 4.
Note that since all images were recorded at about the same height, the z component of the desired motion estimate is rather small after the first reference image and we subsequently ignore it. Instead, we use the distance-to-the-ground measurements to maintain a constant-altitude flight.
- 5.
This threshold depends on the onboard camera’s fov and is selected so as to ensure a significant overlap (more than 80%) between the current camera image and the next reference image.
References
Autonomous flights through image-defined paths (videos). http://mars.cs.umn.edu/research/quadrotor_project.php
Alahi, A., Ortiz, R., Vandergheynst, P.: FREAK: Fast retina keypoint. In: Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition, Providence, RI, pp. 510–517, 16–21 June 2012
Azrad, S., Kendoul, F., Nonami, K.: Visual servoing of quadrotor micro-air vehicle using color-based tracking algorithm. J. Syst. Design Dyn. 4(2), 255–268 (2010)
Bebop Drone. http://www.parrot.com/products/bebop-drone/
Bills, C., Chen, J., Saxena, A.: Autonomous MAV flight in indoor environments using single image perspective cues. In: Proceedings of the IEEE International Conference on Robotics and Automation, Shanghai, China, pp. 5776–5783, 9–13 May 2011
Bourquardez, O., Mahony, R., Guenard, F.C.N.: Image-based visual servo control of the translation kinematics of a quadrotor aerial vehicle. IEEE Trans. Robot. 25(3), 743–749 (2009)
Chaumette, F., Hutchinson, S.: Visual servo control, part i: basic approaches. IEEE Robot. Autom. Mag. 13(4), 82–90 (2006)
Chaumette, F., Hutchinson, S.: Visual servo control, part ii: advanced approaches. IEEE Robot. Autom. Mag. 14(1), 109–118 (2007)
Chen, Z., Birchfield, R.T.: Qualitative vision-based path following. IEEE Trans. Robot. 25(3), 749–754 (2009)
Cormen, T.H., Leiserson, C.E., Rivest, R.L., Stein, C.: Introduction to Algorithms. MIT Press, Cambridge (2001)
Courbon, J., Mezouar, Y., Guenard, N., Martinet, P.: Vision-based navigation of unmanned aerial vehicle. Control Eng. Pract. 18(7), 789–799 (2010)
Courbon, J., Mezouar, Y., Martinet, P.: Indoor navigation of a non-holonomic mobile robot using a visual memory. Auton. Robot. 25(3), 253–266 (2008)
Diosi, A., S̆egvić, S., Remazeilles, A., Chaumette, F.: Experimental evaluation of autonomous driving based on visual memory and image-based visual servoing. IEEE Trans. Robot. 12(3), 870–883 (2011)
Do, T., Carrillo-Arce, L.C., Roumeliotis, S.I.: Autonomous flights through image-defined paths, Technical Report (2015). http://mars.cs.umn.edu/publications.html
Fischler, M., Bolles, R.: Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. Commun. ACM 24(6), 381–395 (1981)
Franklin, G.F., Powell, J.D., Workman, M.L.: Digital Control of Dynamic Systems. Addison-Wesley, Reading (1997)
Goedemé, T., Tuytelaars, T., Gool, L.V., Vanacker, G., Nuttin, M.: Feature based omnidirectional sparse visual path following. In: Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, Alberta, Canada, pp. 1806–1811, 2–6 August 2005
Guo, C.X., Kottas, D.G., DuToit, R.C., Ahmed, A., Li, R., Roumeliotis, S.I.: Efficient visual-inertial navigation using a rolling-shutter camera with inaccurate timestamps. In: Proceedings of the Robotics: Science and Systems Conference, Berkeley, CA, 12–16 July 2014
Hartley, R.I., Zisserman, A.: Multiple View Geometry in Computer Vision, 2nd edn. Cambridge University Press, Cambridge (2004). ISBN: 0521540518
Honegger, D., Meier, L., Tanskanen, P., Pollefeys, M.: An open source and open hardware embedded metric optical flow CMOS camera for indoor and outdoor applications. In: Proceedings of the IEEE International Conference on Robotics and Automation, Karlsruhe, Germany, pp. 1736–1741, 6–16 May 2013
Horn, B.: Closed-form solution of absolute orientation using unit quaternions. J. Opt. Soc. America A 4(4), 629–642 (1987)
Horn, B.: Relative orientation. Int. J. Comput. Vis. 4(1), 59–78 (1990)
Lee, D., Ryan, T., Kim, H.J.: Autonomous landing of a VTOL UAV on a moving platform using image-based visual servoing. In: Proceedings of the IEEE International Conference on Robotics and Automation, Saint Paul, MN, pp. 971–976, 14–18 May 2012
Mariottini, G.L., Roumeliotis, S.I.: Active vision-based robot localization and navigation in a visual memory. In: Proceedings of the IEEE International Conference on Robotics and Automation, Shanghai, China, pp. 6192–6198, 9–13 May 2011
Nguyen, T., Mann, G.K.I., Gosine, R.G.: Vision-based qualitative path-following control of quadrotor aerial vehicle. In: Proceedings of the IEEE International Conference on Unmanned Aircraft Systems, Orlando, FL, pp. 412–417, 27–30 May 2014
Nistér, D.: An efficient solution to the five-point relative pose problem. IEEE Trans. Patter Anal. Mach. Intell. 26(6), 756–770 (2004)
Nistér, D., Stewénius, H.: Scalable recognition with a vocabulary tree. In: Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition, New York, NY, pp. 2161–2168, 17–22 June 2006
Vicon Motion Systems Ltd. http://www.vicon.com/
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG
About this chapter
Cite this chapter
Do, T., Carrillo-Arce, L.C., Roumeliotis, S.I. (2018). Autonomous Flights Through Image-Defined Paths. In: Bicchi, A., Burgard, W. (eds) Robotics Research. Springer Proceedings in Advanced Robotics, vol 2. Springer, Cham. https://doi.org/10.1007/978-3-319-51532-8_3
Download citation
DOI: https://doi.org/10.1007/978-3-319-51532-8_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-51531-1
Online ISBN: 978-3-319-51532-8
eBook Packages: EngineeringEngineering (R0)