• The Journal of Korea Robotics Society
• /
• v.19 no.1
• /
• pp.130-138
• /
• 2024

High-precision 3D Object Detection is a crucial component within autonomous driving systems, with far-reaching implications for subsequent tasks like multi-object tracking and path planning. In this paper, we propose a novel approach designed to enhance the performance of 3D Object Detection, especially in heading angle estimation by employing a moving object segmentation technique. Our method starts with extracting point-wise moving labels via a process of moving object segmentation. Subsequently, these labels are integrated into the LiDAR Pointcloud data and integrated data is used as inputs for 3D Object Detection. We conducted an extensive evaluation of our approach using the KITTI-road dataset and achieved notably superior performance, particularly in terms of AOS, a pivotal metric for assessing the precision of 3D Object Detection. Our findings not only underscore the positive impact of our proposed method on the advancement of detection performance in lidar-based 3D Object Detection methods, but also suggest substantial potential in augmenting the overall perception task capabilities of autonomous driving systems.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.4
• /
• pp.489-496
• /
• 2010

In this paper, a self-organization framework for swarm systems in three dimensions is presented. The framework uses artificial potential functions(APFs) to direct the robots toward the goal as well as to keep them in a swarm system. This research extends conventional APFs used for self-organizations in two dimension environment to three dimensions. In three dimension environment, the ground potential for the boundary surfaces that commonly appear in three dimension environments is proposed. Accordingly, the comparison between the paths without and with the ground potentials shows the necessity and effect of ground potentials. Extensive simulations are given to show the effectiveness of the extended potentials and various properties in three dimension environments.

• 제어로봇시스템학회:학술대회논문집
• /
• 1989.10a
• /
• pp.627-634
• /
• 1989

A collision-free trajectory planning algorithm using the iterative learning concept is proposed for dual robot arms in a 3-D workspace to accurately follow their specified paths with constant velocities. Specifically, a collision-free trajectory minimizing the trajectory error is obtained first by employing the linear programming technique. Then the total operating time is iteratively adjusted based on the maximum trajectory error of the previous iteration so that the collision-free trajectory has no deviation from the specified path and also the operating time is near-minimal.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.10 no.3
• /
• pp.179-188
• /
• 2015

We propose obstacle classification method based on 2D LIDAR(Light Detecting and Ranging) database. The existing obstacle classification method based on 2D LIDAR, has an advantage in terms of accuracy and shorter calculation time. However, it was difficult to classifier the type of obstacle and therefore accurate path planning was not possible. In order to overcome this problem, a method of classifying obstacle type based on width data of obstacle was proposed. However, width data was not sufficient to improve accuracy. In this paper, database was established by width, intensity, variance of range, variance of intensity data. The first classification was processed by the width data, and the second classification was processed by the intensity data, and the third classification was processed by the variance of range, intensity data. The classification was processed by comparing to database, and the result of obstacle classification was determined by finding the one with highest similarity values. An experiment using an actual autonomous vehicle under real environment shows that calculation time declined in comparison to 3D LIDAR and it was possible to classify obstacle using single 2D LIDAR.

• Journal of KIISE:Software and Applications
• /
• v.37 no.8
• /
• pp.641-646
• /
• 2010

This paper presents algorithms for real-time navigation of a humanoid robot with a stereo vision but no other sensors. Using the algorithms, a robot can recognize its 3D environment by retrieving SIFT features from images, estimate its position through the Kalman filter, and plan its path to reach a destination avoiding obstacles. Our approach focuses on estimating the robot’s central walking path trajectory rather than its actual walking motion by using an approximate model. This strategy makes it possible to apply mobile robot localization approaches to humanoid robot localization. Simple collision free path planning and motion control enable the autonomous robot navigation. Experimental results demonstrate the feasibility of our approach.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.49 no.12
• /
• pp.989-996
• /
• 2021

In the case of an unmanned aerial vehicle using a battery as a power source, there are restrictions in performing the mission because the battery capacity is limited. To extend the mission capability, it is important to minimize battery usage while the flight to the mission area. In addition, by using the battery usage prediction model, the possibility of mission completeness can be determined and it can be a criterion for selecting an emergent landing point in the mission planning stage. In this paper, we propose a battery usage prediction model considering as one of the environmental factors in the three-dimensional space. The required power is calculated according to the flight geometry of an unmanned aerial vehicle. True battery usage which is predicted from the required power is verified through the comparison with the battery usage prediction model. The optimal flight trajectory that minimizes battery usage is produced and compared with the shortest travel distance.

• Progress in Medical Physics
• /
• v.28 no.4
• /
• pp.144-148
• /
• 2017

When a high density metallic implant is placed in the path of the proton beam, spatial heterogeneity can be caused due to artifacts in three dimensional (3D) computed tomography (CT) scans. These artifacts result in range uncertainty in dose calculation in treatment planning system (TPS). And this uncertainty may cause significant underdosing to the target volume or overdosing to normal tissue beyond the target. In clinical cases, metal implants must be placed in the beam path in order to preserve organ at risk (OARs) and increase target coverage for tumors. So we should introduce Ti-mesh. In this paper, we measured the lateral dose profile for proton beam using an EBT3 film to confirm dosimetric impact of Ti-mesh when the Ti-mesh plate was placed in the proton beam pathway. The effect of Ti-mesh on the proton beam was investigated by comparing the lateral dose profile calculated from TPS with the film-measured value under the same conditions.

• Journal of the Korean Society for Precision Engineering
• /
• v.13 no.3
• /
• pp.158-164
• /
• 1996

In this paper, the CNC cutting planning module for product with three dimensional solid shape is realized to develop a smart CAD/CAM system which performs systematically from the shape design of procuct by the B-Rep solid modeler to the CNC cutting of product by a machining center. The three dimensional solid shape of product can be easily designed and constructed by the Euler operators and Boolean operators of the solid modeler. And the various functions such as the automatic generation of tool path for the rough and finish cutting processes, the automatic elimination of overcut, the automatic generation of CNC code for the machining center and do on, are established. Especially, the overcut-free tool paths are obtained by splitting the CL solid which is composed of the offset surfaces of the solid shape of product.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.10a
• /
• pp.579-580
• /
• 2016

본 연구에서는 군집 로봇 체계에서 next-best-view(NBV)의 결정과 경로 계획(path planning)을 동시에 수행하는 효율적인 3차원 탐사 알고리즘을 제안한다. NBV 결정 및 경로 계획을 동시에 수행하는 단일 로봇 탐사 알고리즘에는 최근에 제안된 RH-NBVP[1]가 있다. 우리는 각 로봇에게 효율적으로 탐사 영역을 분배하는 Voronoi-biased multi-RRTs(VB-MRRTs)를 제시하며, 이를 통해 군집 로봇 체계에 RH-NBVP를 적용한다. 제시한 VB-MRRTs가 공간을 얼마나 점유하는지 분석하였으며, 3차원 시뮬레이션 상의 군집 로봇 체계에서 VB-MRRTs가 적용된 RH-NBVP의 탐사율 향상을 확인하였다.

• Korean Journal of Computational Design and Engineering
• /
• v.9 no.3
• /
• pp.246-252
• /
• 2004

Presented in this Paper is a Micro Machining Service .(MIMS) based on the World Wide Web technologies. In order to ensure easy access to the service, the web browsers are used as the user interface. The pan geometry as an STL file is uploaded with process parameter for 3-axis CNC milling. Depending on the predefined user level, novice or expert, the user interface requires different parameters for process planning. An STL-based CAM resides in the server and automatically provides NC codes upon user's request. Tool-paths for scanning and pencil-cut, which are interference-free and precise, are created by the curve-based polyhedral machining method. A couple of sample parts were fabricated by a micro endmill with 127 fm diameter. From the tests, the parts fabricated by scanning followed by pencil-cut resulted in less error(within 2%) than the parts machined only by scanning tool-path.