2016
DOI: 10.1016/j.compeleceng.2016.05.003
|View full text |Cite
|
Sign up to set email alerts
|

FPGA design of EKF block accelerator for 3D visual SLAM

Abstract: International audienceThis paper deals with the evaluation of a dedicated architecture to be integrated into an embedded system typically mounted on a micro-aerial vehicle or on smart devices held by an operator. This system performs an Extended Kalman Filter (EKF) based visual odometry (VO) algorithm. An efficient hardware architecture conceived as a systolic array co-processor for EKF loop acceleration is presented. Due to severe limitations in terms of power consumption, real-time performance and physical c… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
2
0

Year Published

2016
2016
2024
2024

Publication Types

Select...
3
3
1

Relationship

0
7

Authors

Journals

citations
Cited by 29 publications
(2 citation statements)
references
References 26 publications
0
2
0
Order By: Relevance
“…As a result, energy-saving research on VSLAM has been conducted in recent years. This includes using FPGA-embedded system designs as the basis for basic VSLAM calculations [ 47 ], analyzing requirements based on dataset efficiency [ 48 ], and selecting related visual algorithms. For the most energy-efficient application, ORB-SLAM2 will be selected, where [ 49 ] mobile robots run on the NVidia Jetson platform, enabling high performance while maintaining low power consumption.…”
Section: Related Workmentioning
confidence: 99%
“…As a result, energy-saving research on VSLAM has been conducted in recent years. This includes using FPGA-embedded system designs as the basis for basic VSLAM calculations [ 47 ], analyzing requirements based on dataset efficiency [ 48 ], and selecting related visual algorithms. For the most energy-efficient application, ORB-SLAM2 will be selected, where [ 49 ] mobile robots run on the NVidia Jetson platform, enabling high performance while maintaining low power consumption.…”
Section: Related Workmentioning
confidence: 99%
“…This could be combined with map management techniques to produce denser maps. Notice that although modern approaches, dominated by the Bundle Adjustement (BA) technique, use high level feature descriptors to solve both detection and matching, and improve the solutions to other problems like place recognition, active search strategies are still active an active field of research in high performance computation specialized works, as mentioned in (Törtei Tertei et al, 2016). Thus, the presented work could be expanded aiming towards highly specific architectures where, through parallelization and other advanced programming techniques, it could be possible to introduce several upgrades, e.g.…”
Section: Discussionmentioning
confidence: 99%