On the VINS Resource-Allocation Problem for a Dual-Camera, Small-Size Quadrotor

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

doi:10.1007/978-3-319-50115-4_47

Cited by 3 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The contributions that are most relevant to our proposal are the one of Davison [37], Lerner et al [38], Mu et al [39], Wu et al [40], and Zhang and Vela [41]. The pioneering work of Davison [37] is one of the first papers to use information theoretic constructs to reason about visual features, and shares many of the motivations discussed in the present paper.…”

Section: Related Work This Work Intersects Several Lines Of Research ...mentioning

confidence: 83%

“…Third, we perform feature selection in a single stage and leverage submodularity to provide formal performance guarantees. Wu et al [40] consider a multi camera system and split the feature selection process into a cascade of two resource-allocation problems: (i) how to allocate resources among the cameras, and (ii) how to select features in each camera, according to the allocated resources. The former problem is solved by taking simplifying assumptions on the distribution of the features, the latter is based on the heuristic feature selection scheme of [42].…”

Section: Related Work This Work Intersects Several Lines Of Research ...mentioning

confidence: 99%

See 1 more Smart Citation

Attention and Anticipation in Fast Visual-Inertial Navigation

2019

View full text Add to dashboard Cite

We study a Visual-Inertial Navigation (VIN) problem in which a robot needs to estimate its state using an on-board camera and an inertial sensor, without any prior knowledge of the external environment. We consider the case in which the robot can allocate limited resources to VIN, due to tight computational constraints. Therefore, we answer the following question: under limited resources, what are the most relevant visual cues to maximize the performance of visual-inertial navigation? Our approach has four key ingredients. First, it is task-driven, in that the selection of the visual cues is guided by a metric quantifying the VIN performance. Second, it exploits the notion of anticipation, since it uses a simplified model for forwardsimulation of robot dynamics, predicting the utility of a set of visual cues over a future time horizon. Third, it is efficient and easy to implement, since it leads to a greedy algorithm for the selection of the most relevant visual cues. Fourth, it provides formal performance guarantees: we leverage submodularity to prove that the greedy selection cannot be far from the optimal (combinatorial) selection. Simulations and real experiments on agile drones show that our approach ensures state-of-the-art VIN performance while maintaining a lean processing time. In the easy scenarios, our approach outperforms appearance-based feature selection in terms of localization errors. In the most challenging scenarios, it enables accurate visual-inertial navigation while appearance-based feature selection fails to track robot's motion during aggressive maneuvers. This paper is conditionally accepted for publication in the IEEE Transaction on Robotics. Please cite the conference version as: L. Carlone and S. Karaman, "Attention and anticipation in fast visual-inertial navigation" 1 This definition oversimplifies the attention mechanisms in humans. While the role of attention is to optimally allocate resources to maximize performance, it is known that some involuntary attention mechanisms can actually hinder the correct execution of a task [2].

show abstract

Section: Related Work This Work Intersects Several Lines Of Research ...mentioning

confidence: 83%

Section: Related Work This Work Intersects Several Lines Of Research ...mentioning

confidence: 99%

Attention and Anticipation in Fast Visual-Inertial Navigation

2019

View full text Add to dashboard Cite

show abstract

Section: Related Workmentioning

confidence: 91%

Attention and anticipation in fast visual-inertial navigation

Carlone

Karaman

2017

2017 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

We study a Visual-Inertial Navigation (VIN) problem in which a robot needs to estimate its state using an on-board camera and an inertial sensor, without any prior knowledge of the external environment. We consider the case in which the robot can allocate limited resources to VIN, due to tight computational constraints. Therefore, we answer the following question: under limited resources, what are the most relevant visual cues to maximize the performance of visual-inertial navigation? Our approach has four key ingredients. First, it is task-driven, in that the selection of the visual cues is guided by a metric quantifying the VIN performance. Second, it exploits the notion of anticipation, since it uses a simplified model for forwardsimulation of robot dynamics, predicting the utility of a set of visual cues over a future time horizon. Third, it is efficient and easy to implement, since it leads to a greedy algorithm for the selection of the most relevant visual cues. Fourth, it provides formal performance guarantees: we leverage submodularity to prove that the greedy selection cannot be far from the optimal (combinatorial) selection. Simulations and real experiments on agile drones show that our approach ensures state-of-the-art VIN performance while maintaining a lean processing time. In the easy scenarios, our approach outperforms appearance-based feature selection in terms of localization errors. In the most challenging scenarios, it enables accurate visual-inertial navigation while appearance-based feature selection fails to track robot's motion during aggressive maneuvers. This paper is conditionally accepted for publication in the IEEE Transaction on Robotics. Please cite the conference version as: L. Carlone and S. Karaman, "Attention and anticipation in fast visual-inertial navigation" IEEE Intl. Conf. on Robotics and Automation (ICRA), pp. 3886-3893, 2017.

show abstract