An Efficient Rescue System with Online Multi-Agent SLAM Framework

Lee, SeungHwan; Kim, HanJun; Lee, Beom-Hee

doi:10.3390/s20010235

Cited by 20 publications

(11 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Specifically, the mAP of the YOLO v2 model is 99%. Nevertheless, based on the class-wise detection results, as indicated in Table 5 , the proposed models outperformed the existing models 9 , 39 , 40 . Besides, although the dataset used is unbalanced, the overall accuracy in the study shows more than 98% to 100% in identifying between the same species and gender of Aedes aegypti , Aedes albopictus , Culex quinquefasciatus (Suppl.…”

Section: Discussionmentioning

confidence: 91%

Deep learning approaches for challenging species and gender identification of mosquito vectors

Kittichai

Pengsakul

Chumchuen

et al. 2021

Sci Rep

View full text Add to dashboard Cite

Microscopic observation of mosquito species, which is the basis of morphological identification, is a time-consuming and challenging process, particularly owing to the different skills and experience of public health personnel. We present deep learning models based on the well-known you-only-look-once (YOLO) algorithm. This model can be used to simultaneously classify and localize the images to identify the species of the gender of field-caught mosquitoes. The results indicated that the concatenated two YOLO v3 model exhibited the optimal performance in identifying the mosquitoes, as the mosquitoes were relatively small objects compared with the large proportional environment image. The robustness testing of the proposed model yielded a mean average precision and sensitivity of 99% and 92.4%, respectively. The model exhibited high performance in terms of the specificity and accuracy, with an extremely low rate of misclassification. The area under the receiver operating characteristic curve (AUC) was 0.958 ± 0.011, which further demonstrated the model accuracy. Thirteen classes were detected with an accuracy of 100% based on a confusion matrix. Nevertheless, the relatively low detection rates for the two species were likely a result of the limited number of wild-caught biological samples available. The proposed model can help establish the population densities of mosquito vectors in remote areas to predict disease outbreaks in advance.

show abstract

Section: Discussionmentioning

confidence: 91%

Deep learning approaches for challenging species and gender identification of mosquito vectors

Kittichai

Pengsakul

Chumchuen

et al. 2021

Sci Rep

View full text Add to dashboard Cite

show abstract

“…In the overlap-based map merging methods [5], [14], [17], map merging is conducted by finding and matching the overlapping areas using additional features extracted by mathematical transform methods such as Hough transformation [15] and Radon transformation [16]. TSD (truncated signed distances) grid-based map merging algorithm [17] has been developed to perform multi-agent TSD-SLAM. Their algorithm was tested in real rescue systems with smart helmets equipped with IMU (inertial measurement unit) and LIDAR (light detection and ranging) sensors.…”

Section: B Related Workmentioning

confidence: 99%

Extended Spectra-Based Grid Map Merging With Unilateral Observations for Multi-Robot SLAM

Lee

2021

IEEE Access

Self Cite

View full text Add to dashboard Cite

This paper deals with the problem of grid map merging in multi-robot SLAM (simultaneous positioning and mapping) where the initial relative pose between robots is unknown. When robots encounter each other, it is easy to obtain a map transformation between robots for grid map merging if bilateral observation measurements are available between robots. However, since the bilateral observation measurements are obtained by encounters between robots, they may limit the availability of using multi-robot systems. To overcome the limitation, spectra-based map merging can be applied without any observation measurements between robots. However, it requires sufficient overlapping areas between indivisual maps of robots, which can also limit the availability of using multi-robot systems. In this paper, therefore, to overcome both limitations, an extension of spectra-based map merging using not bilateral but unilateral observation measurements. The proposed method was tested with datasets obtained from real experiments with mobile robots equipped with a sensor fusion system which can obtain unilateral observation measurements to other robots. Experimental results showed that the proposed map merging method works successfully without any bilateral observation measurements.

show abstract

“…On the one hand, regarding the navigation time in a burning building, researchers in [20] developed a platform to reduce the time with a multi-agent simultaneous localisation and mapping (SLAM) technique. The results indicated that the built map was more accurate than the map obtained using the conventional TSD-SLAM.…”

Section: Location Estimation For Navigation and Firefighter Trackingmentioning

confidence: 99%

HelpResponder—System for the Security of First Responder Interventions

Rodriguez‐Sanchez

Fernández-Jiménez

Jiménez

et al. 2021

Sensors

View full text Add to dashboard Cite

Firefighter’s interventions under dense smoke and flames are hazardous and ideally need an efficient in-advance geo-located actuation plan. The existing communication and sensing technologies should be customized, optimized, and integrated to better know the conditions (flame locations, air condition) before and during the rescue team’s interventions. In this paper, we propose a firefighter intervention architecture, which consists of several sensing devices (flame detectors, carbon dioxide air content) a navigation platform (an autonomous ground wheeled robot), and a communication/localization network (BLE IoT network) that can be used before and during an intervention in rescue or fire extinguishing missions even for indoor or confined spaces. The paper’s key novelty presents our integrated solution, giving some key implementation details and an intensive experimentation campaign in two real firefighter scenarios with real controlled fires. Results carried out in these real indoor scenarios are presented to demonstrate the feasibility of the system. A fire detection system is proposed to improve fire focus in real time and moving in confined spaces with no visibility and physical references. The results obtained in the experimentation show the proposal’s effectiveness in locating the fire focus’s position and orientation reducing time and risk exposure. This kind of location-aware fire integrated systems would significantly impact the speed and security of first responder interventions.

show abstract

An Efficient Rescue System with Online Multi-Agent SLAM Framework

Cited by 20 publications

References 19 publications

Deep learning approaches for challenging species and gender identification of mosquito vectors

Deep learning approaches for challenging species and gender identification of mosquito vectors

Extended Spectra-Based Grid Map Merging With Unilateral Observations for Multi-Robot SLAM

HelpResponder—System for the Security of First Responder Interventions

Contact Info

Product

Resources

About