A5044005499- Human Mobility and Location-Based Analysis
- Indoor and Outdoor Localization Technologies
- Context-Aware Activity Recognition Systems
- Robotics and Sensor-Based Localization
- Mobile Crowdsensing and Crowdsourcing
- Social Robot Interaction and HRI
- Robotics and Automated Systems
- Traffic Prediction and Management Techniques
- Bluetooth and Wireless Communication Technologies
- 3D Surveying and Cultural Heritage
- IoT and Edge/Fog Computing
- Gaze Tracking and Assistive Technology
- Gait Recognition and Analysis
- Time Series Analysis and Forecasting
- Smart Parking Systems Research
- Advanced Manufacturing and Logistics Optimization
- Augmented Reality Applications
- Data Management and Algorithms
- Anomaly Detection Techniques and Applications
- Inertial Sensor and Navigation
- Vehicle License Plate Recognition
- Big Data Technologies and Applications
- AI in Service Interactions
- Spatial and Panel Data Analysis
- Teleoperation and Haptic Systems
Nagoya University
2016-2025
Every year, for ten years now, the IPIN competition has aimed at evaluating real-world indoor localisation systems by testing them in a realistic environment, with movement, using EvAAL framework. The provided unique overview of state-of-the-art systems, technologies, and methods positioning navigation purposes. Through fair comparison performance achieved each system, was able to identify most promising approaches pinpoint critical working conditions. In 2020, included 5 diverse off-site...
Understanding, modeling, and predicting human mobility patterns in urban areas has become a crucial task from the perspectives of traffic disaster risk management, planning, more. HuMob Challenge 2023 aims to predict future movement trajectories based on past 100,000 users[1]. Our team, "uclab2023", considered that model design significantly impacts training prediction times trajectory prediction. To address this, we proposed BERT, commonly used natural language processing, which allows...
As global demand for logistics continues to grow, optimizing the automation and efficiency of distribution warehouse operations is paramount importance. Digitalizing environments, which refers process sensing physical space extracting meaningful information from obtained data, offers a promising solution this challenge. However, converting raw such as video footage captured inside warehouse, into actionable metadata (e.g., tracking movement paths workers products or analyzing usage patterns...
Inspection and repair of road infrastructures are important for safety. While highways motorways periodically inspected with specialized vehicles, the roads which maintained by local governments not because lack budget workforce. In future, however, a large number autonomous driving cars will run everywhere. They equipped laser scanners to recognize surroundings. If we utilize these sensors inspect surface condition automatically, cost inspection be reduced dramatically. this paper, extract...
We have developed an indoor location estimation method using mobile Bluetooth Low Energy (BLE) tags carried by people and BLE scanners fixed to a building. By the method, we can analyze behavior of attendees at some large-scale exhibition, such as order visited booth duration stay. Using has advantages: collect large amount data easily, provide without smart-phones. However, in real environment, signal is unstable due many obstacles. Fingerprinting difficult because arranging booths finishes...
This paper proposes an accurate estimation method of walking speed using deep learning for smartphone-based Pedestrian Dead Reckoning (PDR).PDR requires to estimate and direction pedestrians accurately accelerometer gyroscope.To improve the accuracy PDR, existing works focused key factors (i.e., stride and/or step estimation) by adapting learning.On contrary, our research adapt more directly from sensor data smartphone. We evaluate proposed comparing with conventional PDR method. As a...
There are many inertial sensor based indoor localization methods for smartphone, example, SINS and PDR. However, most of the MEMS sensors smartphones not precise enough these methods. We proposed end-to-end walking speed estimation method using deep learning to perform robust with a low-precision sensor. Currently, we use input data fixed format 200 samples at 100 Hz. sampling rate sequence length should be changed appropriately depending on required accuracy terminal performance. They...
PDR (Pedestrian Dead Reckoning) is a very promising technology for indoor positioning. We held technical challenge, entitled the UbiComp/ISWC 2015 Challenge, consisting of following three categories: algorithm category; Evaluation method and an exhibition. In this paper, we especially focus on several systems category. A skeleton was prepared participants. Using Android skeleton, participants implementing because skeleton's various functions, such as sensor data acquisition, trajectory...
The bases of the approaches UCLab(submission 1) towards SHL recognition challenge are using Random Forest and letting it select important features. Using accelerometer, gyroscope, magnetometer, gravity pressure sensor as input data, features such mean, variance, max, difference max min, main frequency calculated. We find that activities Still, Train, Subway highly similar hard to distinguish. To achieve robust recognition, we make predictions for every segment 3 seconds produce final...
Indoor location estimation has long been researched to realize location-based services. In this paper, we propose an indoor method for Bluetooth Low Energy (BLE) devices using end-to-end LSTM neural network. We focus on large-scale exhibition where is a tough environment wireless due signal strength instability. To achieve higher accuracy, deep learning based methods are proposed rather than trilateration or fingerprint. Existing estimate the from probabilities difference of query and...
Indoor location estimation is essential technology when we analyse the participants' activities in large-scale exhibition. There are some problems with existing methods such as PDR, ultrasound and laser range finder: installation of measurement equipment at large site, cost for equipment, necessity smartphone application. We focus on Bluetooth Low Energy(BLE). Currently, BLE used proximity notification smartphones cannot detect exact distance between two devices. This because radiowave...
Crowd flow forecasting is expected to have a wide range of applications such as human resource allocation, guidance design, marketing, disaster mitigation and congestion prediction for avoiding epidemic COVID-19. challenging because it requires considering both the task capturing temporal dependency data spatial dependence. To address these challenges, in this paper, we propose mechanism referencing time-series features that are important incorporating graph convolution into Transformer,...
The manual creation of a ceiling plan consumes many human resources to confirm the current state existing buildings for renovation. Especially positions and types fixtures are important. In this paper, we propose synthesis method panoramic image using video shot by an omnidirectional camera. We utilize Visual SLAM map frames including ceiling. prioritize clear depiction fixture outlines easier identification fixtures. To draw in resulting image, each is referred from single frame that...
In this paper, LSTM-based neural network is applied to indoor localization using mobile BLE tag's signal strength collected by multiple scanners. Stability of a critical factor wireless for higher accuracy. While traditional methods like trilateration and fingerprinting suffer from noise packet loss, deep learning based perform well. We focus on large-scale exhibition where gets unstable due many people. Proposed consists fully connected layers removal LSTM time-series feature extraction....
This paper proposes an indoor localization method for Bluetooth Low Energy (BLE) devices using end-to-end LSTM neural network. We focus on a large-scale space where there is tough environment wireless due to signal instability. Our proposed adopts localization, which means input time-series of strength and output the estimated location at latest time in input. The network our consists fully-connected layers. use custom-made loss function with 3 error components: MSE, direction travel, leap...
We introduce MetaPo, a mobile robot with spheric display, 360° media I/O and robotic hands for creating unified model of interspace communication. MetaPo works as portal between pairs physical-physical, cyber-cyber cyber-physical spaces to provide 1) panoramic communication multiple remote users, 2) immersive migration mobility functionality. The paper overviews our concept first prototype its hardware software implementation.
We present an assisting system for creating a ceiling plan. Conventional methods of plan are time-consuming and high-cost. Our requires only two inputs from user outputs the panoramic image that shows whole surface. The detects fixtures depicts them seamlessly reliable resulting image. confirmed possibility in with our through experiment.
Spatio-temporal data is utilized in various fields, but its scale continuously growing, leading to significant labor and costs storage processing. Therefore, the value that can be derived from spatio-temporal diluted due management costs. We propose a new flow using metadata common programs for utilization. Traditionally, processing have been implemented processed according each data. defined structure performed based on program. Furthermore, we automated generation of by combining our...