• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.10
• /
• pp.609-617
• /
• 2016

In general, Personal Navigation Systems (PNSs) can be defined systems to acquire pedestrian positional information. GPS is an example of PNS. However, GPS can only be used where the GPS signal can be received. Pedestrian Dead Reckoning (PDR) can estimate the positional information of pedestrians using Inertial Measurement Unit (IMU). Therefore, PDR can be used for GPS-disabled areas. This paper proposes a PDR scheme considering various movement types over GPS-disabled areas as combat environments. We propose a movement distance estimation scheme and movement direction estimation scheme as pedestrian's various movement types such as walking, running and crawling using IMU. Also, we propose a fusion algorithm between GPS and PDR to mitigate the lack of accuracy of positional information at the entrance to the building. The proposed algorithm has been tested in a real test bed. In the experimental results, the proposed algorithms exhibited an average position error distance of 5.64m and position error rate in goal point of 3.41% as a pedestrian traveled 0.6km.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.2
• /
• pp.115-121
• /
• 2022

As city environments have recently become crowded, there are many obstacles that interfere with the walking of the visually impaired on pedestrian roads. Typical examples include ballads, parking breakers and standing signs, which usually do not get in the way, but blind people may be injured by collisions. To solve such a problem, many solutions have been proposed, but they are limited in applied in practical environments due to the several restrictions such as outside use only, inaccurate obstacle sensing and requirement of special devices. In this paper, we propose a new method to automatically detect obstacles while walking on the pedestrian roads and warn the collision risk in advance by using only sensors embedded in typical mobile phones. The proposed method supports the walking of the visually impaired by notifying the type of obstacles appearing in front of them as well as the distance remaining from the obstacles. To accomplish this goal, we utilized an object recognition technology applying the latest deep learning algorithms in order to identify the obstacles appeared in real-time videos. In addition, we also calculate the distance to the obstacles using the number of steps and the pedestrian's stride. Compared to the existing walking support technologies for the visually impaired, our proposed method ensures efficient and safe walking with only simple devices regardless of the places.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2010.05a
• /
• pp.755-756
• /
• 2010

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.7 no.2
• /
• pp.19-26
• /
• 2013

This paper propose a surface EMG signal based gait phase recognition method that selects features and channels adaptively. The proposed method can be used to control powered artificial prosthetic for lower limb amputees and can reduce overhead in real-time pattern recognition by selecting adaptive channels and features in an embedded device. The method can enhance the classification accuracy by adaptively selecting channels and features based on sensitivity and specificity of each subject because EMG signal patterns may vary according to subject's locomotion convention. In the experiments, we found that the muscles with highest recognition rate are different between human subjects. The results also show that the average accuracy of the proposed method is about 91% whereas those of existing methods using all channels and/or features is about 50%. Therefore we assure that sEMG signal based gait phase recognition using small number of adaptive muscles and corresponding features can be applied to control powered artificial prosthetic for lower limb amputees.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.16 no.6
• /
• pp.1-15
• /
• 2017

Recently terrorism targets unspecified masses and causes massive destruction, which is so-called Super Terrorism. Many countries have tried hard to protect their citizens with various preparation and safety net. With inexpensive and advanced technologies of sensors, the surveillance systems have been paid attention, but few studies associated with the classification of the pedestrians' trajectories and the difference among themselves have attempted. Therefore, we collected individual trajectories at Samseoung Station using an analytical solution (system) of pedestrian trajectory by LiDAR sensor. Based on the collected trajectory data, a comprehensive framework of classifying the types of pedestrians' trajectories has been developed with data normalization and "trajectory association rule-based algorithm." As a result, trajectories with low similarity within the very same cluster is possibly detected.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.11 no.11
• /
• pp.4087-4094
• /
• 2010

The sensor data which is measured by Quadruped robot is utilized to recognize the physical environment or other information and to control the posture and walking of robot system. In order to control the robot precisely, high accuracy of sensor data is required, most of these sensors however, belongs to expensive and low-durable products. Moreover, these are exposed excessive load operation in a field condition if it is applied to field robot system. This issue becomes more serious one when the robot system is manufactured as a mass product. As in this context, this study suggests a virtual sensor technology to alternate or assist the main sensor system. This scheme is realized by using back-propagation algorithm of neural network theory, and the quality of estimated sensor data could be improved through the algorithmic and hardware based treatments. This study performs the various trial to identify the effective parameters which effect to the quality and reliability of estimated sensor data and tries to show the possibility of proposed methodology.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.34 no.6
• /
• pp.659-665
• /
• 2010

This paper proposes a control algorithm for quadruped robots moving on irregularly sloped uneven surfaces. Since the body balance of a quadruped robot is controlled by the forces acting on its feet during touchdown, the ground reaction force (GRF) is controlled for stable running. The desired GRF for each foot is generated on the basis of the desired galloping pattern; this GRF is then compared with the actual contact force. The difference between the two forces is used to modify the foot trajectory. The desired force is realized by considering a combination of the rate change of the angular and linear momenta at flight. Then, the amplitude of the GRF to be applied at each foot in order to achieve the desired linear and angular momenta is determined by fuzzy logic. Dynamic simulations of galloping motion were performed using RecurDyn; these simulations show that the proposed control method can be used to achieve stable galloping for a quadruped robot on irregularly sloped uneven surfaces.

• Journal of Korea Spatial Information System Society
• /
• v.12 no.1
• /
• pp.76-84
• /
• 2010

Many pedestrian simulation models for micro-scale spaces as building indoor areas have been proposed for the last decade and two models - social force model and floor field model - are getting attention. Among these, CA-based floor field model is viewed more favourable for computer simulations than computationally complex social force model. However, Kirchner's floor field model has limitations in capturing the differences in dynamic values of different agents and this study proposes an enhanced algorithm. This study improved the floor field model in order for an agent to be able to exclude the influences of its own dynamic values by changing the data structure, and, also modified the initial dynamic value problem in order to fit more realistic environment. In the simulations, real 3D building data stored in a spatial DBMS were used considering future integration with indoor localization sensors and real time applications.

• Journal of Digital Policy
• /
• v.2 no.4
• /
• pp.33-39
• /
• 2023

With a continuous occurrence of right-turn traffic accidents at intersections, there is an increasing demand for measures to address these incidents. In response, a technology has been developed to detect the presence of pedestrians through object detection in CCTV footage at right-turn areas and display warning messages on the screen to alert drivers. The YOLO (You Only Look Once) model, a type of object detection model, was employed to assess the performance of object detection. An algorithm was also devised to address misidentification issues and generate warning messages when pedestrians are detected. The accuracy of recognizing pedestrians or objects and outputting warning messages was measured at approximately 82%, suggesting a potential contribution to preventing right-turn accidents

• The KIPS Transactions:PartB
• /
• v.19B no.3
• /
• pp.177-182
• /
• 2012

We propose effective motion and observation models for the position of a WiFi-equipped smartphone user in large indoor environments. Three component motion models provide better proposal distribution of the pedestrian's motion. Our Gaussian interpolation-based observation model can generate likelihoods at locations for which no calibration data is available. These models being incorporated into the particle filter framework, our WiFi fingerprint-based localization algorithm can track the position of a smartphone user accurately in large indoor environments. Experiments carried with an Android smartphone in a multi-story building illustrate the performance of our WiFi localization algorithm.