Accuracy enhancement for the front-end tracking algorithm of RGB-D SLAM

Hu, Fuwen; Cheng, Jingli; Bao, Yunchang; He, You

doi:10.1007/s11370-019-00299-2

Cited by 5 publications

(3 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…B. Fang's research group proposes a dynamic scene SLAM algorithm in view of boundary box and depth continuity; This algorithm uses a deep bounding box to fill in pixels for random search, and eliminates the influence of the target through dynamic feature filtering; The experimental results show that the positioning accuracy and real-time performance of this method in complex dynamic scenes meet the design expectations [7]. F. Hu et al proposed an improved ORB-SLAM front-end tracking algorithm that utilizes a uniform velocity model to track effective frames and adjacent frames, and matches similar frames; Experimental data shows that this method can increase the number of effective tracking frames and reduce the computational complexity by two times [8]. J. Dong and his team members proposed an improved RGB-D SLAM scheme that utilizes ORB for feature point extraction and descriptor calculation, and matches the current frame (CF) with the map; The research results show that this method reduces the Root-mean-square (RMS) deviation by 9% on average, and improves the indexing effect of point cloud images [9] This study indicates that robot vision and cruising can improve the safety of vehicle autonomous driving [12].…”

Section: Related Workmentioning

confidence: 76%

Improved ORB-SLAM2 Mobile Robot Vision Algorithm Based on Multiple Feature Fusion

Hu,

Zhu,

Wang

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Traditional wheeled robot vision algorithms suffer from low texture tracking failures. Therefore, this study proposes a vision improvement algorithm for mobile robots in view of multi feature fusion; This algorithm introduces line surface features and Manhattan Frame on the basis of traditional algorithms, and proposes an improved algorithm in view of multi-sensor fusion to improve tracking accuracy. The experiment shows that the average Root-mean-square deviation of the position of the improved mobile robot vision algorithm in view of multi feature fusion is 0.02 in nine data packets of the Tum dataset; The average Root-mean-square deviation of the position of the data packet successfully tracked by the traditional wheeled robot vision algorithm is 0.016; It improved the average accuracy by 11.11%, which is 31.03% higher than the average accuracy of the Manhattan wheeled robot vision algorithm. Compared to the multi feature fusion based vision improvement algorithm for mobile robots and the closed-loop detection based multi-sensor improvement algorithm, the accuracy of the closed-loop detection based multi-sensor improvement algorithm has increased by 0.655% and 10.47%, respectively. The outcomes indicate that the improved algorithm can improve the accuracy of mobile robot tracking, thereby expanding its application range. INDEX TERMSMulti feature fusion, Mobile robots, Visual algorithms, Multi sensor fusion, Encoder

show abstract

Section: Related Workmentioning

confidence: 76%

Improved ORB-SLAM2 Mobile Robot Vision Algorithm Based on Multiple Feature Fusion

Hu,

Zhu,

Wang

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Hu et al [20] consider a depth RGB camera (RGB-D) for visual SLAM and present a method to improve the algorithm shown by Mur-Artal and Tardós [40], minimizing the tracking losses due to pure rotation, sudden movements and noise. In experimental results, the proposed method shows improvements in the tracking accuracy with low computational costs.…”

Section: Perception Localization and Filteringmentioning

confidence: 99%

An integrated solution for an autonomous drone racing in indoor environments

Rezende

Miranda

Rezeck

et al. 2021

Intel Serv Robotics

View full text Add to dashboard Cite

The use of drones is becoming more present in modern daily life. One of the most challenging tasks associated with these vehicles is the development of perception and autonomous navigation systems. Competitions such as Artificial Intelligence Robotic Racing (AIRR) and Autonomous Drone Racing were launched to drive the advances in such strategies, requiring the development of integrated systems for autonomous navigation in a racing environment. In this context, this paper presents an improved integrated solution for autonomous drone racing, which focuses on simple, robust, and computationally efficient techniques to enable the application in embedded hardware. The strategy is divided into four modules: (i) A trajectory planner computes a path that passes through the sequence of desired gates; (ii) a perception system that obtains the global pose of the vehicle by using an onboard camera; (iii) a localization system which merges several sensed information to estimate the drone's states; and, (iv) an artificial vector field-based controller in charge of following the plan by using the estimated states. To evaluate the performance of the entire integrated system, we extended the FlightGoggles simulator to use gates similar to those used in the AIRR competition. A computational cost analysis demonstrates the high performance of the proposed perception method running on limited hardware commonly embedded in aerial robots. In addition, we evaluate the localization system by comparing it with ground truth and when there is a disturbance in the location of the gates. Results in a representative race circuit based on the AIRR competition showed the proposed strategy's competing performance, presenting itself as a feasible solution for drone racing systems.

show abstract

“…In conjunction with the aforementioned hardware, the navigation system integrates simultaneous localization and mapping (SLAM) technology to attain accurate positioning within localized areas [4]. A standard SLAM positioning system typically comprises two primary components: the front-end [5], tasked with sensing and generating equipment and software that characterizes environmental features while calculating pose information for the intelligent entity; and the back-end [6], encompassing software designed to estimate positioning errors and optimize navigation algorithms. The amalgamation of these devices and software facilitates precise positioning of intelligent agents.…”

Section: Introductionmentioning

confidence: 99%

Air-ground multi-agent system cooperative navigation based on factor graph optimization SLAM

Liu,

Wang,

Liu

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

In this study, a cooperative navigation algorithm centered on Factor Graph Optimization - Simultaneous Localization and Mapping (FGO-SLAM) is presented for an air-ground multi-agent system. The algorithm prioritizes the control of error statuses during the position and attitude estimation procedure throughout the entire back-end optimization process. In the conventional Extended Kalman Filtering (EKF) algorithm, periodic cumulative errors may arise, introducing uncertainty to the estimation process. The application of the factor graph optimization algorithm not only mitigates deviation but also stabilizes errors, thereby eliminating the accumulation of periodic errors. In comparison to the practical EKF-SLAM, FGO-SLAM serves as a semi-offline optimization system that leverages key frames to minimize computational load. During multi-agent simulations, when two or more agents have overlapping field views, landmark data is merged, enhancing the optimization effectiveness. Through simulation experiments, the proposed algorithm demonstrates a 40% reduction in position error and a 41% reduction in attitude error, affirming the efficacy of FGO-SLAM for cooperative navigation.

show abstract

Accuracy enhancement for the front-end tracking algorithm of RGB-D SLAM

Cited by 5 publications

References 28 publications

Improved ORB-SLAM2 Mobile Robot Vision Algorithm Based on Multiple Feature Fusion

Improved ORB-SLAM2 Mobile Robot Vision Algorithm Based on Multiple Feature Fusion

An integrated solution for an autonomous drone racing in indoor environments

Air-ground multi-agent system cooperative navigation based on factor graph optimization SLAM

Contact Info

Product

Resources

About