|
Fast Location Survey of DGNSS Reference Station to Support UAV Navigation |
|
Dong-Kyeong Lee, Jiyun Lee(KAIST, Republic of Korea) |
| Mobile DGNSS formed with multiple UAVs can support missions on a battlefield by providing navigation with high accuracy and integrity. For time efficiency and maneuverability of the system, the reference stations that are deployed and sited on the battlefield should be able to determine their precise position rapidly before generating the differential corrections. This paper presents a methodology for fast location surveying of the reference stations, which improves the GNSS-based position estimates of reference stations using precise measurements of relative distances between the stations. |
|
|
MDP-Based Mission Planning for Multi-UAV Persistent Surveillance |
|
Byeong-Min Jeong(Korea Aerospace Industry, Republic of Korea), Jung-Su Ha, Han-Lim Choi(KAIST, Republic of Korea) |
| This paper presents a methodology to generate task flow for conducting a surveillance mission using multiple UAVs, when the goal is to persistently maintain the uncertainty level of surveillance regions as low as possible. The mission planning problem is formulated as a Markov decision process (MDP), which is a infinite-horizon discrete stochastic optimal control formulation and often leads to a periodic task flows to be implemented in a persistent manner. Numerical simulations verify the applicability of the proposed decision scheme. |
|
|
Optimal Path Planning Based on Spline-RRT* for Fixed-Wing UAVs Operating in Three-Dimensional Environments |
|
Dasol Lee, HanJun Song, David Hyunchul Shim(KAIST, Republic of Korea) |
| This paper proposes a spline-RRT* algorithm and describes its application to path planning for fixed-wing UAVs operating in three-dimensional environments. The tree structure of the proposed spline-RRT* algorithm is extended by using a spline method based on a cubic Bézier curve. Through the use of this spline method, the algorithm can produce a smooth path without any post-processing and can also handle the initial approach direction, that is, the heading and flight path angle for the target UAV. A dynamic feasibility check, incorporating a geometric collision check, is also run as part of th |
|
|
Ad-hoc Network for Inter-Vehicle Communication of Multiple UAVs |
|
Youngjoo Kim, Hyochoong Bang(KAIST, Republic of Korea) |
| This paper addresses a method to implement an ad-hoc network for inter-vehicle communication of multiple UAVs. During cooperative missions, it is essential to share information between different UAVs. The ad-hoc network provides greater autonomy to the team of UAVs because it is free of intervention of ground stations. The challenges we consider for implementing inter-vehicle communication are including increased data amount, distinguishing data, and system complexity. This paper provides the solutions to these problems in terms of hardware and software. |
|
|
An indoor autonomous flight of multiple ornithopters following a circular path |
|
Ho-Young Kim, Jun-Seong Lee, Jae-Hung Han(KAIST, Republic of Korea) |
| This work demonstrates an autonomous flight of multiple ornithopters in an indoor flight environment. The tested ornithopters have no on-board sensor for flight state estimation or feedback control. Instead, the indoor flight environment with a motion capture system provides flight state variables by tracking the markers on the ornithopters. A feedback controller is designed for each ornithopter to follow a circular path with a desired altitude and radius. To avoid collision, the angular speed of each ornithopter is controlled by changing the desired radius of each ornithopter. |
|
|
Design of a Multiple-UAS Mission Model using Activity-based Modeling |
|
Hyunjin Park, Seongsik Jeong, Jaemyung Ahn(KAIST, Republic of Korea) |
| This paper introduces an activity-based modeling design and subsequently analyzes mission models of a UAS mission with multiple UA. We use activity-based modeling within the Arena simulation environment as our main modeling methodology, where each stage of a mission’s progress is represented by process models and the UA performing said missions correspond to the agents. All processes are connected to create a single comprehensive model that simulates the behavior of the UA agents as they perform missions... |
|