무차별 대입법과 원 경로를 이용한 지형 추종의 궤적 생성에 관한 연구

강승훈,이현주,이상철,전지연,임동주 한국항공우주학회 2023 韓國航空宇宙學會誌 Vol.51 No.10

적의 방공망에 피탐되지 않기 위한 방법 중 하나로 지형 추종이 사용될 수 있다. 지형 추종을 수행하기위해 비행 궤적을 생성할 필요가 있다. 항공기 진행 방향의 지형 프로파일을 사용하여 최소 이격거리, 항공기의 속력, 상승각 및 하강각 제한 그리고 수직가속도의 제한을 만족하는 비행 궤적을 생성한다. 본 논문에서는 기존에 개발된 지형 추종 궤적 생성 알고리즘에 적용할 수 있는 새로운 방법을 제안한다. 지형추종 궤적 생성 알고리즘은 두 단계로 구성될 수 있다. 먼저, 지형 프로파일이 입력되면 최소 이격거리와비행 상승각 및 하강각의 제한을 만족하는 비행 경로를 생성한다. 비행 경로를 생성하기 위한 기존 방법은모폴로지 연산을 사용한다. 다음으로, 비행 경로에 항공기의 속력, 비행 상승각 및 하강각 제한 그리고 수직가속도의 제한을 만족하는 원 경로를 생성하여 적용하고 궤적으로 변환한다. 비행 경로 생성에 소요되는시간을 단축하기 위해 모폴로지 연산 방법의 대안으로 무차별 대입법을 제안한다. 지형 추종 궤적 생성 알고리즘은 일정 주기마다 궤적을 생성할 때 이전 궤적과 현재 궤적이 연결되지 않는 궤적 불연속 문제가 발생하는 경우가 있다. 궤적 불연속 문제를 해결하기 위해 반지름 제한을 완화하여 원 경로를 생성하는 방법을 제안하며 기존 원 경로 생성 방법을 보완하였다. 무차별 대입법과 모폴로지 연산 방법을 이용하여 비행 경로를 생성하고 생성 시간을 비교하였고, 무차별 대입법 기반의 비행 경로에 보완한 원 경로 방법을 적용하여 궤적 불연속 문제가 발생하지 않는 것을 확인하였다. One of the methods to avoid detection by enemy air defense systems is terrain following. To perform terrain following, it is necessary to generate flight trajectories. This paper proposes new methods that can be applied to existing terrain following trajectory generation algorithms. The terrain following trajectory generation algorithm consists of two steps. First, when the terrain profile is given, a flight path which satisfies the minimum clearance height and limits of climb and descent angles is generated. The generation of flight paths is conducted by using the morphology method in the existing method. Next, a circular path is applied to the flight path, and then converted into a trajectory. The circular path satisfies the aircraft's speed, limits on climb and descent angles and the vertical acceleration. In this paper, to reduce the runtime of generating the flight path, a brute force method is proposed. The terrain following trajectory generation algorithm sometimes generates the discontinued trajectories which refers that the previous and current trajectories are not connected. To address this problem, a method to generate circular path without the limitation of radius is proposed, which could improve the existing circular path method. Through simulation, runtimes of the flight path generation using the brute force method and the morphology method are compared. Additionally, the improved circular path method is applied to the flight path using the brute force. The result of the improved circular path shows that the trajectory discontinuity problem did not occur

노동경력 배열분석을 통해 본 코호트별 계급궤적 비교

남춘호(Nahm, Choon-Ho) 한국지역사회학회 2016 지역사회연구 Vol.24 No.2

This article aims to analyze the pattern of intra-generational class mobility and compare the type of class trajectory of each cohort through sequence analysis, for which class sequences between ages 20 to 40 has been constituted from the work history data of the 17<SUP>th</SUP> wave KLIPS. Class trajectories of the X generation, especially petty bourgeois class trajectory, unskilled worker trajectory, and sales/service worker trajectory, show a low persistence and stability compared to the class trajectories of the baby boomer generation, which is estimated to reflect the flexibility of labor market since the financial crisis. De-industrialization and the transition to service society seem to lead to the increase of professional and semi-professional trajectory in the upper class and service worker trajectory in the lower class in recent generations. The impact of the transition into service society is different by gender. Whereas highly educated professional trajectory is surging for men, semi-professional trajectory, sales/service worker trajectory, and clerical and service workers with career breaks in non-employment trajectory are growing rapidly for women. More than a half of the female work sequences are classified as non-employment class trajectory with career breaks even among generation X, which indicates that career break is still the most important factor that separates the pattern of female class trajectories.

쿼드로터 무인기를 위한 궤적 생성 및 추종 제어 시스템 개발

이다솔,심현철 제어·로봇·시스템학회 2019 제어·로봇·시스템학회 논문지 Vol.25 No.5

In this paper, we describe the development of a trajectory generation and trajectory tracking control system for a quadrotor UAV. The trajectory generation algorithm generates minimum snap trajectories which minimizes effort for actuation of motors, and the geometric control based trajectory tracking controller is utilized for tracking given trajectories precisely. The trajectory tracking controller generates a thrust command and angular rate commands for an inner-loop angular rate controller. To identify flight control performance, we developed a quadrotor platform and carried out flight experiments using circle and figure 8 trajectories. The flight experiment results show that the quadrotor UAV precisely follows the commanded trajectories in real flight environments.

시계열 생성적 적대 신경망을 이용한 비행체 궤적 합성 데이터 생성 및 비행체 궤적 예측에서의 활용에 관한 연구

박인희(In Hee Park),이창진(Chang Jin Lee),정찬호(Chanho Jung) 한국전기전자학회 2021 전기전자학회논문지 Vol.25 No.4

딥러닝을 포함한 머신러닝 기법을 기반으로 비행체의 궤적 설계, 제어, 최적화, 예측 등의 작업을 수행하기 위해서는 일정한 양 이상의 비행체 궤적 데이터를 필요로 한다. 그러나 다양한 이유(예를 들어 비행체 궤적 데이터셋 구축에 필요한 비용, 시간, 인력 등)로 일정한 양 이상의 비행체 궤적 데이터를 확보하기 어려운 경우가 존재한다. 이러한 경우 합성 데이터 생성이 머신러닝을 가능하게 하는 방법 중 하나가 될 수 있다. 본 논문에서는 이와 같은 가능성을 탐구하기 위하여 시계열 생성적 적대 신경망을 이용하여 비행체 궤적 합성 데이터를 생성하고 평가하였다. 또한 비행체의 상태를 인식하기 위한 비행체 궤적 예측 작업에서 합성 데이터의 활용 가능성을 탐구하기 위하여 다양한 ablation study(비교 실험)를 수행하였다. 본 논문에서 제시된 생성 평가 및 비교 실험 결과는 비행체 궤적 합성 데이터 생성 및 비행체 궤적 관련 작업에서 합성 데이터의 활용 가능성에 대한 연구를 수행하고자 하는 연구자들에게 실질적인 도움이 될 것으로 예상한다. In order to perform tasks such as design, control, optimization, and prediction of flight vehicle trajectories based on machine learning techniques including deep learning, a certain amount of flight vehicle trajectory data is required. However, there are cases in which it is difficult to secure more than a certain amount of flight vehicle trajectory data for various reasons. In such cases, synthetic data generation could be one way to make machine learning possible. In this paper, to explore this possibility, we generated and evaluated synthetic flight vehicle trajectory data using time-series generative adversarial neural network. In addition, various ablation studies (comparative experiments) were performed to explore the possibility of using synthetic data in the aircraft trajectory prediction task. The experimental results presented in this paper are expected to be of practical help to researchers who want to conduct research on the possibility of using synthetic data in the generation of synthetic flight vehicle trajectory data and the work related to flight vehicle trajectories.

Design of Trajectory Generator for Performance Evaluation of Navigation Systems

Jae Hoon Son,Sang Heon Oh,Dong-Hwan Hwang 사단법인 항법시스템학회 2023 Journal of Positioning, Navigation, and Timing Vol.12 No.4

In order to develop navigation systems, simulators that provide navigation sensors data are required. A trajectory generator that simulates vehicle motion is needed to generate navigation sensors data in the simulator. In this paper, a trajectory generator for evaluating navigation system performance is proposed. The proposed trajectory generator consists of two parts. The first part obtains parameters from the motion scenario file whereas the second part generates position, velocity, and attitude from the parameters. In the proposed trajectory generator six degrees of freedom, halt, climb, turn, accel turn, spiral, combined, and waypoint motions are given as basic motions with parameters. These motions can be combined to generate complex trajectories of the vehicle. Maximum acceleration and jerk for linear motion and maximum angular acceleration and velocity for rotational motion are considered to generate trajectories. In order to show the usefulness of the proposed trajectory generator, trajectories were generated from motion scenario files and the results were observed. The results show that the proposed trajectory generator can accurately simulate complex vehicle motions that can be used to evaluate navigation system performance.

실시간 계층적 심층강화학습 기반 드론 궤적 생성 알고리즘 파라미터 제어

지창훈,한연희,문성태 한국통신학회 2023 韓國通信學會論文誌 Vol.48 No.10

With the increasing use of drones, research on drone trajectory generation algorithms has gained significant momentum. These algorithms aim to generate real-time trajectories while considering obstacle avoidance. Recent advancements have shown promising results in generating safe and efficient trajectories in complex dynamic environments, such as forests, as well as controlling multiple drones simultaneously. However, most existing drone trajectory generation algorithms impose limitations on the maximum speed and acceleration parameters to ensure drone stability. These restrictions on speed-related parameters hinder the efficiency and practicality of drones. In this paper, we propose a novel approach called “Hierarchical Deep Reinforcement Learning-Based Active Parameter Control Algorithm” that addresses this limitation. This algorithm dynamically sets the maximum speed and acceleration parameters of a drone based on the real-time environment using a hierarchical reinforcement learning framework. The upper layer agent in the hierarchy is responsible for adjusting the maximum speed and acceleration parameters considering the current environmental conditions. The lower layer agent then utilizes these parameters to generate a real-time trajectory. Notably, this approach can be applied to all drone trajectory generation algorithms that involve setting maximum speed and maximum acceleration. Through extensive simulations, we demonstrate that applying the proposed algorithm to drone trajectory generation algorithms results in superior performance in terms of speed, path length, and path smoothness. These improvements showcase the potential of our approach in enhancing the efficiency and overall capabilities of drones operating in complex and dynamic environments.

안정적인 ZMP 궤적을 추정하는 2족 로봇의 궤적 생성

황성욱(Sung Wook Hwang),조재욱(Jae Uk Cho),박종현(Jong Hyeon) 대한기계학회 2013 대한기계학회 춘추학술대회 Vol.2013 No.12

In this paper, the trajectory generation method for a biped robot to follow a stable ZMP trajectory is proposed. In order to generate the walking trajectory, the robot is modeled as one particle. And the momentum equation with respect to the desired ZMP(Zero Moment Point) of the robot is used. The desired ZMP of the simplified model should be fixed at supporting foot during a single support phase and moves from the current supporting foot to next supporting foot. An initial position and last position of the desired ZMP are used to generate the trajectory for the double support phase. During the double support phase, the desired ZMP trajectory is generated by using Null space to follow the stable ZMP trajectory. The biped robot can walk stably by following the stable ZMP trajectory through the proposed trajectory generation method. The effectiveness of the proposed method was proved through simulations.

대용량 GPS 궤적 데이터를 위한 효율적인 클러스터링

김태용(Taeyong Kim),박보국(Bokuk Park),박진관(Jinkwan Park),조환규(Hwan-Gue Cho) Korean Institute of Information Scientists and Eng 2016 정보과학회논문지 Vol.43 No.1

Digital road map generation is primarily based on artificial satellite photographing or in-site manual survey work. Therefore, these map generation procedures require a lot of time and a large budget to create and update road maps. Consequently, people have tried to develop automated map generation systems using GPS trajectory data sets obtained by public vehicles. A fundamental problem in this road generation procedure involves the extraction of representative trajectory such as main roads. Extracting a representative trajectory requires the base data set of piecewise line segments(GPS-trajectories), which have close starting and ending points. So, geometrically similar trajectories are selected for clustering before extracting one representative trajectory from among them. This paper proposes a new divide- and-conquer approach by partitioning the whole map region into regular grid sub-spaces. We then try to find similar trajectories by sweeping. Also, we applied the Fréchet distance measure to compute the similarity between a pair of trajectories. We conducted experiments using a set of real GPS data with more than 500 vehicle trajectories obtained from Gangnam-gu, Seoul. The experiment shows that our grid partitioning approach is fast and stable and can be used in real applications for vehicle trajectory clustering.

다자유도 기계 시스템의 모션 제어를 위한 동기된 온라인 궤적 생성 방법

원대희(Daehee Won),신은철(Eunchol Shin) 대한전기학회 2017 전기학회논문지 Vol.66 No.9

This work presents a synchronized on-line trajectory generation algorithm for mechanical system with multiple degrees-of-freedom. Proposed algorithm is designed to generate time-optimized trajectories and synchronized trajectories under the constraints such as maximum speed, acceleration, deceleration, non-zero initial velocity, etc. Also, because of small computation time, therefore this can be applied in real-time and it is easier to change the trajectory when an event occurs. We verified the feasibility through various trajectory generation.

Trajectory Generation of a Moving Object for a Mobile Robot in Predictable Environment

TaeSeok Jin,JangMyung Lee 한국정밀공학회 2004 International Journal of Precision Engineering and Vol.5 No.1

In the field of machine vision using a single camera mounted on a mobile robot, although the detection and tracking of moving objects from a moving observer, is complex and computationally demanding task. In this paper, we propose a new scheme for a mobile robot to track and capture a moving object using images of a camera. The system consists of the following modules: data acquisition, feature extraction and visual tracking, and trajectory generation. And a single camera is used as visual sensors to capture image sequences of a moving object. The moving object is assumed to be a point-object and projected onto an image plane to form a geometrical constraint equation that provides position data of the object based on the kinematics of the active camera. Uncertainties in the position estimation caused by the point￢object assumption are compensated using the Kalman filter. To generate the shortest time trajectory to capture the moving object, the linear and angular velocities are estimated and utilized. The experimental results of tracking and capturing of the target object with the mobile robot are presented.