High-Accuracy Robot Indoor Localization Scheme Based on Robot Operating System Using Visible Light Positioning

Guan, Weipeng; Chen, Shihuan; Wen, Shangsheng; Tan, Zequn; Song, Hongzhan; Hou, Wenyuan

doi:10.1109/jphot.2020.2981485

Cited by 100 publications

(53 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The location estimate (x,ŷ) of the above-described analytical multilateration method can finally be obtained with least squares optimization [40]. The solution adopted in this work employs a non-linear least squares approach, using the Broyden-Fletcher-Goldfarb-Shanno algorithm (BFGS) [41], where the objective is to minimize the cost function J(x,ŷ) given by Equation (20) using Equation (19) with p the number of available equations.…”

Section: Relative Rss Based Multilateration Schemementioning

confidence: 99%

“…The fingerprinting method is based on matching measured data with the contents of a large database that consists of calibration measurements to obtain a location estimate. Depending on the VLP method, the sensor that is selected as a receiver generally is a single Photodiode (PD), an array of photodiodes or because of good anti-interference properties a camera/image sensor [19,20].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Experimental Evaluation of Machine Learning Methods for Robust Received Signal Strength-Based Visible Light Positioning

Raes

Knudde

Bruycker

et al. 2020

Sensors

View full text Add to dashboard Cite

In this work, the use of Machine Learning methods for robust Received Signal Strength (RSS)-based Visible Light Positioning (VLP) is experimentally evaluated. The performance of Multilayer Perceptron (MLP) models and Gaussian processes (GP) is investigated when using relative RSS input features. The experimental set-up for the RSS-based VLP technology uses light-emitting diodes (LEDs) transmitting intensity modulated light and a single photodiode (PD) as a receiver. The experiments focus on achieving robustness to cope with unknown received signal strength modifications over time. Therefore, several datasets were collected, where per dataset either the LEDs transmitting power is modified or the PD aperture is partly obfuscated by dust particles. Two relative RSS schemes are investigated. The first scheme uses the maximum received light intensity to normalize the received RSS vector, while the second approach obtains RSS ratios by combining all possible unique pairs of received intensities. The Machine Learning (ML) methods are compared to a relative multilateration implementation. It is demonstrated that the adopted MLP and GP models exhibit superior performance and higher robustness when compared to the multilateration strategies. Furthermore, when comparing the investigated ML models, the GP model is proven to be more robust than the MLP for the considered scenarios.

show abstract

Section: Relative Rss Based Multilateration Schemementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Experimental Evaluation of Machine Learning Methods for Robust Received Signal Strength-Based Visible Light Positioning

Raes

Knudde

Bruycker

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…However, this strategy may not work when the robot is operated in dark and narrow environments. Guan et al [ 16 ] designed a visible light communication localization and navigation package based on the distributed framework of the robot operating system (ROS), which contains the basic operation control of the robots, LED identification detection, and dynamic cm-level visible light positioning algorithms for robot indoor localization. Lv et al [ 17 ] presented a novel indoor localization method for a skid-steering mobile robot by fusing readings from an encoder, gyroscope, and magnetometer, which can be read as an enhanced dead-reckoning localization method.…”

Section: Related Workmentioning

confidence: 99%

Fusion of the SLAM with Wi-Fi-Based Positioning Methods for Mobile Robot-Based Learning Data Collection, Localization, and Tracking in Indoor Spaces

Lee

Moon

Lee

et al. 2020

Sensors

View full text Add to dashboard Cite

The ability to estimate the current locations of mobile robots that move in a limited workspace and perform tasks is fundamental in robotic services. However, even if the robot is given a map of the workspace, it is not easy to quickly and accurately determine its own location by relying only on dead reckoning. In this paper, a new signal fluctuation matrix and a tracking algorithm that combines the extended Viterbi algorithm and odometer information are proposed to improve the accuracy of robot location tracking. In addition, to collect high-quality learning data, we introduce a fusion method called simultaneous localization and mapping and Wi-Fi fingerprinting techniques. The results of the experiments conducted in an office environment confirm that the proposed methods provide accurate and efficient tracking results. We hope that the proposed methods will also be applied to different fields, such as the Internet of Things, to support real-life activities.

show abstract

“…Currently, most VLP systems require multiple LOS links to implement accurate localization. Most of them [25], [27] have an average positioning error between 3cm to 7cm within a positioning range between 2m to 3m. The VLP system presented in this paper has an average positioning error of 13cm at a distance of 3m.…”

Section: Backgrounds and Motivationsmentioning

confidence: 99%

Visual-Inertial Fusion Based Positioning Systems

Zhang

Kane

2020

IEEE Access

View full text Add to dashboard Cite

In this paper, we developed a visible light positioning (VLP) system using a camera and lowcost inertial measurement units (IMUs). Applying computer vision and sensor fusion techniques, our VLP system is able to estimate the angle of arrival (AoA) and the distance from a landmark to a mobile device. Due to the complementary nature between IMUs and cameras, we are able to improve the performance of VLP systems by applying sensor fusion. Currently, most optical positioning systems require at least two line-of-sight (LOS) links, so the coverage is not always satisfactory. Using a single round light-emitting diode (LED) panel or two coplanar black thick rings as the landmark, our VLP system only needs one LOS link to estimate the orientation and position of the mobile device. By activating inertial navigation, our VLP system is able to perform localization even if the landmark is temporarily blocked by obstacles. We derived approximated upper bounds of the angular errors and applied visual-inertial sensor fusion in estimating the Euler angles of the mobile device. Since the weights of sensor fusion are determined by upper bounds, the expected maximum errors are minimized in our positioning system. In our field experiments, the positioning system has an average positioning error of 0.18m with an effective positioning range of 7m. Compared to similar types of positioning systems, our system has significant improvements in positioning range without sacrificing positioning accuracy.

show abstract

High-Accuracy Robot Indoor Localization Scheme Based on Robot Operating System Using Visible Light Positioning

Cited by 100 publications

References 24 publications

Experimental Evaluation of Machine Learning Methods for Robust Received Signal Strength-Based Visible Light Positioning

Experimental Evaluation of Machine Learning Methods for Robust Received Signal Strength-Based Visible Light Positioning

Fusion of the SLAM with Wi-Fi-Based Positioning Methods for Mobile Robot-Based Learning Data Collection, Localization, and Tracking in Indoor Spaces

Visual-Inertial Fusion Based Positioning Systems

Contact Info

Product

Resources

About