Autonomous Flights Through Image-Defined Paths

Do, Tien Van; Carrillo-Arce, Luis C.; Roumeliotis, Stergios I.

doi:10.1007/978-3-319-51532-8_3

Cited by 5 publications

(6 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…the five dof transformation degenerates into a three dof rotation-only transformation (see Do et al, 2015, for more details) between I t and I r k . This three dof transformation can be computed (see Appendix C.1) by employing either (2pt RANSAC) two pairs of feature correspondences, satisfying…”

Section: Technical Approachmentioning

confidence: 99%

“…At this point, we should note that the motion estimate from equation (1), and potentially equation (2), is not reliable when the baseline between I t and I k r is short. In particular, when the distance between two images is significantly shorter than the distance to the feature from either image (i.e., d I k normalr I t << I t d f i , I k r d f i ) the five dof transformation degenerates into a three dof rotation-only transformation (see Do et al, 2015, for more details) between I t and I k r . This three dof transformation can be computed (see Appendix C.1) by employing either (2pt RANSAC) two pairs of feature correspondences, satisfying…”

Section: Technical Approachmentioning

confidence: 99%

“…Note also that when information (e.g., from ultrasound sensors) about nearby obstacles is available, we can modify r so that the quadrotor can smoothly avoid obstacles while maintaining its course (see Do et al, 2015, for more details). After finalizing the desired motion r , we provide it to the proportional–integral–derivative (PID) controller to compute the control actions.…”

Section: Technical Approachmentioning

confidence: 99%

“…A preliminary version of this paper was presented in Do et al (2015), where we introduced the concept of using both wide-baseline and short-baseline RANSACs and demonstrated the robustness of this approach when navigating through tight spaces. A limitation of Do et al (2015), however, was that, owing to the high processing requirements of the 5pt RANSAC (45 ms per image pair), the navigation algorithm could only run at 8 Hz on the quadrotor's processor. To address this issue in this work, we employ the gravity direction in the 3pt + 1 RANSAC to increase the navigation algorithm's speed to 15 Hz.…”

Section: Introduction and Related Workmentioning

confidence: 99%

See 3 more Smart Citations

High-speed autonomous quadrotor navigation through visual and inertial paths

Carrillo-Arce

Roumeliotis

2018

The International Journal of Robotics Research

Self Cite

View full text Add to dashboard Cite

This paper addresses the problem of autonomous quadrotor navigation within indoor spaces. In particular, we focus on the case where a visual map of the area, represented as a graph of linked images, is constructed offline (from visual and potentially inertial data collected beforehand) and used to determine visual paths for the quadrotor to follow. In addition, during the actual navigation, the quadrotor employs both wide-and short-baseline random sample consensuses (RANSACs) to efficiently determine its desired motion toward the next reference image and handle special motions, such as rotations in place. In particular, when the quadrotor relies only on visual observations, it uses the 5pt and 2pt algorithms in the wide-and short-baseline RANSACs, respectively. On the other hand, when information about the gravity direction is available, significant gains in speed are realized by using the 3pt + 1 and 1pt + 1 algorithms instead. Lastly, we introduce an adaptive optical-flow algorithm that can accurately estimate the quadrotor's horizontal velocity under adverse conditions (e.g., when flying over dark, textureless floors) by progressively using information from more parts of the images. The speed and robustness of our algorithms are evaluated experimentally using a commercial-off-the-shelf quadrotor navigating in the presence of dynamic obstacles (i.e., people walking) along lengthy corridors and through tight corners, as well as across building floors via poorly lit staircases.

show abstract

Section: Technical Approachmentioning

confidence: 99%

Section: Technical Approachmentioning

confidence: 99%

Section: Technical Approachmentioning

confidence: 99%

Section: Introduction and Related Workmentioning

confidence: 99%

See 2 more Smart Citations

High-speed autonomous quadrotor navigation through visual and inertial paths

Carrillo-Arce

Roumeliotis

2018

The International Journal of Robotics Research

Self Cite

View full text Add to dashboard Cite

show abstract

“…Several ways to automate the flight of drones have been proposed, including using sensors [42], camera images [9], inputting waypoints as a file [5] or automating the controls [33]. Image based autonomous flights have been described [18] and the use of a platform for the verification of image based navigation explored [35]. Landmark based visual navigation for use in cases of GPS failure has also been investigated [6], along with using network connections to control fleet of drones [46].…”

Section: Related Workmentioning

confidence: 99%

DIMPL: a distributed in-memory drone flight path builder system

Shukla¹,

Chen

2018

J Big Data

View full text Add to dashboard Cite

As use of drones rapidly expands, it is aided by improvements in technology such as high speed cameras, sensors, and processors able to analyze the data rapidly and efficiently on the drone. Better scalability in processing image, terrain, weather and surface data and using it to aid in navigation has also allowed them to become increasingly autonomous during flight. Drones have proven very useful in both military battlefield and civilian tasks. Along with common civilian tasks in education [30], studying natural phenomena [45], reconnaissance [36] and conservation [28] they have been used increasingly in surveying farms [41], forests [10] and borders [27]. Drones can either be controlled manually by an operator or fly autonomously, but as the use for an operator increases the cost considerably, it is clearly preferable for them to operate autonomously. Autonomous flight presents challenges in terrain navigation, however, as a multitude of flight path scenarios such as variations in altitude and the density of objects to be surveyed must be taken into account. Flight planning

show abstract