Kyle O'Keefe

Dr. Kyle O'Keefe

orcid

Research Focus & Metrics

Academic Career Length

29Years (1998-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 April 2018 in Sensors
Verified
Journal Article
Open access
PDF file
PGPaul Verlaine Gakne
Paul Verlaine Gakne
KOKyle O'Keefe
Kyle O'Keefe
This paper presents a method of fusing the ego-motion of a robot or a land vehicle estimated from an upward-facing camera with Global Navigation Satellite System (GNSS) signals for navigation purposes in urban environments. A sky-pointing camera is mounted on the top of a car and synchronized with a GNSS receiver. The advantages of this configuration are two-fold: firstly, for the GNSS signals, the upward-facing camera will be used to classify the acquired images into sky and non-sky (also known as segmentation). A satellite falling into the non-sky areas (e.g., buildings, trees) will be rejected and not considered for the final position solution computation. Secondly, the sky-pointing camera (with a field of view of about 90 degrees) is helpful for urban area ego-motion estimation in the sense that it does not see most of the moving objects (e.g., pedestrians, cars) and thus is able to estimate the ego-motion with fewer outliers than is typical with a forward-facing camera. The GNSS and visual information systems are tightly-coupled in a Kalman filter for the final position solution. Experimental results demonstrate the ability of the system to provide satisfactory navigation solutions and better accuracy than the GNSS-only and the loosely-coupled GNSS/vision, 20 percent and 82 percent (in the worst case) respectively, in a deep urban canyon, even in conditions with fewer than four GNSS satellites.

29 February 2020 in Sensors
Verified
Journal Article
Open access
PDF file
SNSharareh Naghdi
Sharareh Naghdi
KOKyle O'Keefe
Kyle O'Keefe
One of the popular candidates in wireless technology for indoor positioning is Bluetooth Low Energy (BLE). However, this technology faces challenges related to Received Signal Strength Indicator (RSSI) fluctuations due to the behavior of the different advertising channels and the effect of human body shadowing among other effects. In order to mitigate these effects, the paper proposes and implements a dynamic Artificial Intelligence (AI) model that uses the three different BLE advertising channels to detect human body shadowing and compensate the RSSI values accordingly. An experiment in an indoor office environment is conducted. 70% of the observations are randomly selected and used for training and the remaining 30% are used to evaluate the algorithm. The results show that the AI model can properly detect and significantly compensate RSSI values for a dynamic blockage caused by a human body. This can significantly improve the RSSI-based ranges and the corresponding positioning accuracies.

19 June 2018 in Sensors
Verified
Journal Article
Open access
PDF file
APAli Pirsiavash
Ali Pirsiavash
ABAli Broumandan
Ali Broumandan
GLGérard Lachapelle
Gérard Lachapelle
KOKyle O'Keefe
Kyle O'Keefe
Various measurement monitoring techniques are investigated to mitigate the effect of global navigation satellite systems (GNSS) code multipath through error correction, stochastic weighting of measurements and detection and exclusion (or de-weighting) of affected measurements. Following a comprehensive review of each approach, the paper focuses on detection/exclusion and detection/de-weighting techniques where several single and dual-frequency monitoring metrics are employed in a combination with time-averaging and the M of N detection strategy. A new Geometry-Free (GF) detection metric is proposed given its capability to be combined with a preceding Code-Minus-Carrier (CMC)-based error correction to reduce the number of excluded or de-weighted measurements and thus preserve the measurement geometry. Three geometry-based algorithms, namely measurement subset testing, consecutive exclusion and iterative change of measurement weights are investigated to address multipath scenarios with multiple simultaneously affected measurements. Experimental results are provided using GPS L1, L2C and L5 data collected in multipath environments for static and kinematic scenarios. For GPS L1, the proposed combined method shows more than 38% improvement over a conventional Carrier-to-Noise-density ratio (C/N0)-based Least-Squares (LS) solution in all but deep urban canyons. Lower performance was observed for L2C and L5 frequencies with a limited number of satellites in view.

1 December 2009 in Journal of Navigation
Verified
Journal Article
Closed access
PDF file
GMGlenn MacGougan
Glenn MacGougan
KOKyle O'Keefe
Kyle O'Keefe
RKRichard Klukas
Richard Klukas
Ultra-wideband (UWB) ranging radios, an emerging technology that offers precise, short distance range measurements are investigated as a method to augment carrier-phase GPS positioning. A commercially available UWB ranging system is used in a tightly-coupled GPS and UWB real-time kinematic (RTK) system. The performance of the tightly-coupled system is evaluated in static and kinematic testing. This work demonstrates that UWB errors can be successfully estimated in a real-time filter. The results of static testing show that the integrated solution provides better accuracy, better ability to resolve integer ambiguities and enhanced fixed ambiguity solution availability compared with GPS alone. In kinematic testing in a degraded GPS environment, sub-decimetre accuracy was maintained.

30 July 2009 in Measurement Science and Technology
Verified
Journal Article
Closed access
GMGlenn MacGougan
Glenn MacGougan
KOKyle O'Keefe
Kyle O'Keefe
RKRichard Klukas
Richard Klukas

30 October 2018 in IEEE Communications Letters
Verified
Journal Article
Closed access
ZHZhe He
Zhe He
YLYou Li
You Li
LPLing Pei
Ling Pei
KOKyle O'Keefe
Kyle O'Keefe
With the recent advances of technology innovation in the Internet-of-Things (IoT) era, radio chips are able to transmit over long distances with extremely low energy consumption. While extending the range of communication links, the ability to provide large scale location-based services (LBS) solutions using native physical layer (PHY) parameters from IoT networks will dramatically widen the availability of IoT applications. This letter proposes an enhanced Gaussian process based localization (EGPL) solution for such a low power wide area network (LPWAN). It effectively deals with intermittent signals over a large area caused by low communication throughput, interference or packet collisions in LPWAN. Furthermore, a parametric model enhancement combining indoor and outdoor hypotheses and signal propagation statistics is proposed and evaluated. Field tests over a 37,500 square meter area have been conducted. Results show that the proposed method can provide LBS-quality positioning with a 2D root mean square (RMS) error of 20 to 30 meters, with an accuracy improvement of 29.8% outdoors and 40.6% indoors respectively compared to the traditional fingerprinting method.