A Benchmark Comparison of Monocular Visual-Inertial Odometry Algorithms for Flying Robots

Delmerico, Jeffrey A.; Scaramuzza, Davide

doi:10.1109/icra.2018.8460664

Cited by 338 publications

(243 citation statements)

References 31 publications

Supporting

Mentioning

241

Contrasting

Order By: Relevance

“…On a static track (Figure 5a), a SLAMbased state estimator [5], [11] would have less drift than a VIO baseline, but we empirically found the latency of existing open-source SLAM pipelines to be too high for closed-loop control. A benchmark comparison of latencies of monocular visual-inertial SLAM algorithms for flying robots can be found in [50].…”

Section: B Experiments In Simulationmentioning

confidence: 99%

Deep Drone Racing: From Simulation to Reality With Domain Randomization

et al. 2020

Self Cite

View full text Add to dashboard Cite

Dynamically changing environments, unreliable state estimation, and operation under severe resource constraints are fundamental challenges that limit the deployment of small autonomous drones. We address these challenges in the context of autonomous, vision-based drone racing in dynamic environments. A racing drone must traverse a track with possibly moving gates at high speed. We enable this functionality by combining the performance of a state-of-the-art planning and control system with the perceptual awareness of a convolutional neural network (CNN). The resulting modular system is both platform-and domain-independent: it is trained in simulation and deployed on a physical quadrotor without any fine-tuning. The abundance of simulated data, generated via domain randomization, makes our system robust to changes of illumination and gate appearance. To the best of our knowledge, our approach is the first to demonstrate zero-shot sim-to-real transfer on the task of agile drone flight. We extensively test the precision and robustness of our system, both in simulation and on a physical platform, and show significant improvements over the state of the art.

show abstract

Section: B Experiments In Simulationmentioning

confidence: 99%

Deep Drone Racing: From Simulation to Reality With Domain Randomization

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…The state-of-the-art EVIO system, UltimateSLAM [43], operates by independently tracking visual features from pseudo-images reconstructed from events and optional images from a conventional camera, and fusing the tracks with inertial measurements using an existing optimization backend [23]. Here, we go one step further and directly apply an off-the-shelf VIO system (specifically, VINS-Mono [37], which is state-of-the-art [11]) to videos reconstructed from events using either our approach, MR, or HF, and evaluate against UltimateSLAM. As is standard [55,39,43], we use sequences from the Event Camera Dataset [31], which contain events, frames, and IMU measurements from a DAVIS240C [7] sensor.…”

Section: Visual-inertial Odometrymentioning

confidence: 99%

Events-To-Video: Bringing Modern Computer Vision to Event Cameras

Rebecq

Ranftl

Koltun

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Self Cite

250

202

View full text Add to dashboard Cite

Event cameras are novel sensors that report brightness changes in the form of asynchronous "events" instead of intensity frames. They have significant advantages over conventional cameras: high temporal resolution, high dynamic range, and no motion blur. Since the output of event cameras is fundamentally different from conventional cameras, it is commonly accepted that they require the development of specialized algorithms to accommodate the particular nature of events. In this work, we take a different view and propose to apply existing, mature computer vision techniques to videos reconstructed from event data. We propose a novel recurrent network to reconstruct videos from a stream of events, and train it on a large amount of simulated event data. Our experiments show that our approach surpasses state-of-the-art reconstruction methods by a large margin (> 20%) in terms of image quality. We further apply off-the-shelf computer vision algorithms to videos reconstructed from event data on tasks such as object classification and visual-inertial odometry, and show that this strategy consistently outperforms algorithms that were specifically designed for event data. We believe that our approach opens the door to bringing the outstanding properties of event cameras to an entirely new range of tasks. A video of the experiments is available at https://youtu.be/IdYrC4cUO0I

show abstract

“…We use VINS-Mono [11], a tightly-coupled visual-inertial odometry framework that has been shown to perform favorably when compared to other state of the art open source state estimation algorithms [2]. VINS-Mono jointly optimizes vehicle motion, feature locations, camera-IMU extrinsics, and IMU biases over a sliding window of monocular images and preintegrated IMU measurements.…”

Section: State Estimationmentioning

confidence: 99%

Fast and Agile Vision-Based Flight with Teleoperation and Collision Avoidance on a Multirotor

Spitzer

Yang

Yao

et al. 2020

Springer Proceedings in Advanced Robotics

View full text Add to dashboard Cite

We present a multirotor architecture capable of aggressive autonomous flight and collision-free teleoperation in unstructured, GPS-denied environments. The proposed system enables aggressive and safe autonomous flight around clutter by integrating recent advancements in visual-inertial state estimation and teleoperation. Our teleoperation framework maps user inputs onto smooth and dynamically feasible motion primitives. Collision-free trajectories are ensured by querying a locally consistent map that is incrementally constructed from forward-facing depth observations. Our system enables a non-expert operator to safely navigate a multirotor around obstacles at speeds of 10 m/s. We achieve autonomous flights at speeds exceeding 12 m/s and accelerations exceeding 12 m/s 2 in a series of outdoor field experiments that validate our approach.Authors are with the

show abstract

A Benchmark Comparison of Monocular Visual-Inertial Odometry Algorithms for Flying Robots

Cited by 338 publications

References 31 publications

Deep Drone Racing: From Simulation to Reality With Domain Randomization

Deep Drone Racing: From Simulation to Reality With Domain Randomization

Events-To-Video: Bringing Modern Computer Vision to Event Cameras

Fast and Agile Vision-Based Flight with Teleoperation and Collision Avoidance on a Multirotor

Contact Info

Product

Resources

About