Localization and Place Recognition Using an Ultra-Wide Band (UWB) Radar

Takeuchi, Eijiro; Elfes, Alberto; Roberts, Jonathan

doi:10.1007/978-3-319-07488-7_19

Cited by 16 publications

(6 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Place recognition with Ultra Wide Band (UWB) radar is presented in [17] by matching received signals to a database of waveforms which represent signatures of places. Although the UWB class of radar is capable of very high update rates, it is shown in [18] that the FMCW class is superior in raw measurement quality, measured maximum range, and worstcase precision.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Kidnapped Radar: Topological Radar Localisation using Rotationally-Invariant Metric Learning

Săftescu

Gadd

Martini

et al. 2020

2020 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

This paper presents a system for robust, large-scale topological localisation using Frequency-Modulated Continuous-Wave (FMCW) scanning radar. We learn a metric space for embedding polar radar scans using CNN and NetVLAD architectures traditionally applied to the visual domain. However, we tailor the feature extraction for more suitability to the polar nature of radar scan formation using cylindrical convolutions, anti-aliasing blurring, and azimuth-wise max-pooling; all in order to bolster the rotational invariance. The enforced metric space is then used to encode a reference trajectory, serving as a map, which is queried for nearest neighbours (NNs) for recognition of places at run-time. We demonstrate the performance of our topological localisation system over the course of many repeat forays using the largest radar-focused mobile autonomy dataset released to date, totalling 280 km of urban driving, a small portion of which we also use to learn the weights of the modified architecture. As this work represents a novel application for FMCW radar, we analyse the utility of the proposed method via a comprehensive set of metrics which provide insight into the efficacy when used in a realistic system, showing improved performance over the root architecture even in the face of random rotational perturbation.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Section I), the FMCW class is therefore our preferred sensor. Furthermore, evaluation in [17] was performed in indoor and forested environments, whereas our work is deployed in built environments representative of urban driving.…”

Section: Related Workmentioning

confidence: 99%

Kidnapped Radar: Topological Radar Localisation using Rotationally-Invariant Metric Learning

Săftescu

Gadd

Martini

et al. 2020

2020 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

show abstract

“…a ground-based technology that sends and receives. Carrier-less radio impulses using extremely accurate timing [34][35][36][37][38][39]. Compared with other wireless technologies, such as Wi-Fi, ZigBee and Bluetooth, it has high precision and is less affected by external occlusion [40].…”

Section: Introductionmentioning

confidence: 99%

Development and Testing of a New UWB Positioning Measurement Tool to Assist in Forest Surveys

et al. 2022

View full text Add to dashboard Cite

Forest resource inventory is a significant part of the sustainable management of forest ecosystems. Finding methods to accurately estimate the diameter at breast height (DBH), tree height and tree position is a significant part of forest resource inventory. The traditional methods of forest resource inventory are expensive, difficult, laborious and time-consuming; the existing systems are not convenient to carry, resulting in low working efficiency. In addition, it is usually necessary to rely on a forest compass, DBH taper and RTK or handheld GPS to set up the plot. These instruments each have a single function and cannot achieve accurate positioning under the forest canopy. Therefore, it is necessary to update the existing equipment and technology. This study aimed to design. a multi-functional, high-precision, real-time. positioning intelligent tree-measuring instrument that integrates plot the set-up, DBH measurement, tree height measurement and tree position measurement. The instrument is based on the ultra-wideband positioning principle, sensor technology, image processing technology, trigonometric functions, tree surveying and other related theories and realizes the functions of plot set-up, tree position measurement, DBH measurement, tree height measurement and other functions. The device was tested in four square plots. The results showed that the root mean squared. error (RMSE). of the tree position estimates ranged from 0.07 m to 0.16 m, while the relative root mean squared error (rRMSE) of the DBH estimates of individual trees ranged from 3.01 to 6.43%, which is acceptable for practical applications in traditional forest inventory. The rRMSE of the tree height estimates ranged from 3.47 to 5.21%. Furthermore, the cost of this instrument is only about one-third that of traditional forestry survey tools, while the work efficiency is three times that of the traditional measurement methods. Overall, the results confirmed that the tree measuring instrument is a practical tool for obtaining. accurate measurements of the tree position, DBH and tree height for forest inventories.

show abstract

“…By converting the signals generated by pulse-echo radar measurements to spectrograms, it is possible to produce unique fingerprints of locations in the environment, which can then be used by RatSLAM to successfully map the surroundings and localize the mobile agent within this map. This takes the technique of directly using signal waveforms as labels, as seen in [20], one step further by increasing the use of the information contained in the frequency spectrum of the echo. In addition to our own novel SLAM setup, we execute an existing SLAM system in parallel which serves as ground truth.…”

Section: Introductionmentioning

confidence: 99%

RadarSLAM: Biomimetic SLAM using ultra-wideband pulse-echo radar

Schouten

Steckel

2017

2017 International Conference on Indoor Positioning and Indoor Navigation (IPIN)

View full text Add to dashboard Cite

This paper presents a novel method for using an ultra-wideband (UWB), super high frequency (SHF) pulse-echo radar sensor as a biomimetic sensing mechanism to successfully solve the Simultaneous Localization and Mapping (SLAM) problem. Due to recent advances in consumer radar technology it has become possible to sample the received echo signals well above their Nyquist frequency. This Nyquist-conform sampling permits the conversion of the signal waveforms into spectrograms which contain spatiospectral cues caused by the interactions of the echoes with features of the environment and the antenna's radiation pattern. Such spectrograms can therefore serve as distinct labels for individual locations, allowing for the identification and recognition of these locations. By adapting an existing acoustic SLAM system (BatSLAM) we have developed a system that demonstrates the feasibility of our proposed method; the results validate the potential of using pulse-echo radar as an exteroceptive sensory modality in topological SLAM systems.

show abstract

Localization and Place Recognition Using an Ultra-Wide Band (UWB) Radar

Cited by 16 publications

References 10 publications

Kidnapped Radar: Topological Radar Localisation using Rotationally-Invariant Metric Learning

Kidnapped Radar: Topological Radar Localisation using Rotationally-Invariant Metric Learning

Development and Testing of a New UWB Positioning Measurement Tool to Assist in Forest Surveys

RadarSLAM: Biomimetic SLAM using ultra-wideband pulse-echo radar

Contact Info

Product

Resources

About