|
Radar Velocity Estimation Using RANSAC,SNR based Error Compensation Algorithm |
|
Minho Lee, Hoang Viet Do, Kyeong-Wook Seo, Dongyun Hwang, Jin Woo Song(Sejong University, Korea) |
| This paper addresses method for compensating radar velocity measurement. The Random Sample Consensus(RANSAC) algorithm is utilized for outlier removal, and the Signal-to-Noise Ratio (SNR) parameter is employed as a weighting factor to estimate and compensate for relative velocity measurements of multiple targets acquired from the radar. To validate the proposed approach, experiments were conducted using an Unmanned Ground Vehicle (UGV) in a real building environment. The experimental results demonstrate that the method enhances radar velocity measurements more accurately, as evidenced by the experimental results. |
|
|
DOPP Study |
|
Sangkyung Sung, Donggyun Kim(Konkuk University, Korea) |
| In this paper, we study how path planning that reflects the sensor characteristics of three-dimensional LiDAR can reduce the likelihood of SLAM failing to estimate position and attitude and distorting the map. Typical path planning algorithms are designed with the assumption that the robot's position during travel is known exactly. However, real-world robot pathfinding relies on navigation algorithms such as SLAM to estimate position. Therefore, a problem with the navigation algorithm can have a catastrophic effect on the navigation system. In this paper, we present a method for analyzing the observability of geometry in three-dimensional point cloud maps to avoid areas with high potential f |
|
|
Robust Walking Direction Estimation via Principal Component Pursuit in Pedestrian Dead Reckoning |
|
Jae Wook Park, Jae Hong Lee, Chan Gook Park(Seoul National University, Korea) |
| In this paper, we address the misalignment problem in the pedestrian walking direction estimation by focusing on the PCA-based method. While this method exploits the advantages of PCA, it is also subject to its limitations, notably its susceptibility to outliers due to abnormal actions beyond normal walking, which are commonly encountered in PDR scenarios. To ensure robustness against outliers, we propose a novel PCA-based method for estimating the walking direction. Through experiments conducted on various actions, we evaluate the performance of the proposed method, demonstrating that our proposed method achieves a performance improvement compared to classical PCA. |
|
|
HNN-Transformer Integrated Network for Estimating Robot Position |
|
Chang Ho Kang(Kumoh National Institute of Technology, Korea), Sun Young Kim(Kunsan National University, Korea) |
| This paper proposes a novel model architecture, the Hamiltonian neural network (HNN)-Transformer, which capitalizes on the strengths of both Hamiltonian neural networks and Transformers to effectively model and predict the behavior of physical systems. The proposed structure is designed to solve the problem that arises over time when predicting the state of a 2D robotic system. The HNN component of our model makes it possible to incorporate known physical laws into the learning process, while the Transformer component enables effective processing of sequentially input data. The performance of the proposed method was confirmed through simulation. |
|
|
Performance evaluation of quasi-optimal multi-constellation satellite selection methods |
|
Taek Geun Lee, Seok Ho Lee, Yun Seo Choi, Hyung Keun Lee(Korea Aerospace University, Korea) |
| Positioning accuracy by Global Navigation Satellite Systems (GNSSs) is influenced by the number of visibility satellites as well as the geometric placement of the satellites. To achieve feasible accuracy, the optimal satellite selection method is not appropriate for real-time applications due to large computational burden. To solve this problem, extensive studies have been performed and several efficient quasi-optimal methods are available currently. This study compares the three most representative quasi-optimal satellite selection methods in terms of accuracy and computational burden. The comparison will be performed with respect to the open-sky and semi-open-sky environments. |
|
|
Development of an Advanced Navigation System for Autonomous Mobile Robots for Logistics Environments |
|
Jun-Hyeon Choi, Sang-Hyeon Bae, Ye-Chan An, Tae-Yong Kuc(Sungkyunkwan University, Korea) |
| In this paper, we proposes the development of a novel navigation system for Autonomous Mobile Robots (AMR) in a logistics environment.The proposed navigation system includes methods for optimal path planning for single or multiple robots in a warehouse environment, improved localization techniques, and obstacle detection methods. By utilizing three maps, the system plans specialized paths for logistics warehouses, and addresses the limitations of conventional probability-based localization methods by utilizing semantic features extracted from the radar sensor for position recognition. |
|