:: ICCAS 2015 ::

WC08		[OS] Navigation and Control Application for Military Robot Systems
Time : 16:30~18:00 Room : 203		Chair : Yong Woon Park (ADD, )



16:30~16:45	WC08-1

Whole-Body Motion Control for a Rescue Robot using Robust Task-Priority based CLIK
Wonsuk Lee, Youngwoo Lee, Gyuheyun Park, Seongil Hong, Byunghun Choi, Younsik Kang(Agency for Defense Develoment, Korea, Republic of )

This paper introduces a rescue robot to extract a casualty from hazardous environment and suggests a specific whole-body behavior strategy suitable for the rescue mission taking into account the characteristics of the robot configuration. Since the robot has redundant degrees of freedom, a task-priority based CLIK as differential inverse kinematics approach is adopted for real-time implementation and the redundant domain is utilized to reflect some restrictions in terms of safety and stability of the robot.



16:45~17:00	WC08-2

Optimization of Multiple Sensor Data Pipeline for Real-time 3D Terrain Reconstruction
Seoungjae Cho, Seongjo Lee, Kyhyun Um, Kyungeun Cho(Dongguk University-Seoul, Korea, Republic of ), Sungdae Sim, Yong Woon Park(Agency of Defense Development, Korea, Republic of )

Remote-control technology is required in an unmanned vehicle such that it can replace humans for executing tasks in various extreme environments. In particular, a remote scene must be reconstructed using 3D meshes for enabling a user to remotely control an unmanned vehicle easily and intuitively. To this end, a large amount of multiple sensor data would be processed using various algorithms in real time. But it is difficult to implement 3D terrain reconstruction in high-quality and real time. This paper proposes the optimization of the architecture of a multiple-sensor-data pipeline.



17:00~17:15	WC08-3

Satellite and Street Map Matching Method Using Iterative Closest Point
Kang Jeong Min, Park Jin Bae(Yonsei Universiity, Korea, Republic of )

This paper proposes a satellite and street maps matching by using Iterative Closest Point (ICP) algorithm. The proposed method is two data point sets matching extracted from maps. First, we extract the feature points from a map. Second, we consist of input data set from feature points, and then take the center of mass transformation process. Finally, map matching is performed with two data point sets. The ICP algorithm is used as a matching method. The error results of the rotation and translation will be compared with original input values.



17:15~17:30	WC08-4

Vertical Channel Damping Loop Design using Vertical Position and Velocity Information
TOK SON CHOE(Yonsei University, Korea, Republic of ), Won Sang Ra(Handong Global University, Korea, Republic of ), Jin Bae Park(Yonsei University, Korea, Republic of ), Kwanghoon Kim, Myonghwan Ahn(LIG Nex1 Co., Ltd., Korea, Republic of )

In this paper, a vertical channel damping loop design using vertical position and velocity information is newly proposed. The proposed method uses vertical position and velocity information acquired by additional sensors. The Kalman filter is constructed to calculate vertical channel damping loop gains. The proposed method is simulated by using INS and additional sensors error statistics. Results demonstrate the usefulness and practicality of the proposed method.



17:30~17:45	WC08-5

Localization of UGV Based on 2-Point Absolute Pose Algorithm
Sung-In Choi, Pathum Rathnayaka, Soon-Yong Park(Kyungpook National University, Korea, Republic of )

This paper introduces a new approach to recognize the position of a UGV utilizing a Digital Surface Map along with forward direction three dimensional distance sensors and cameras. The proposed method consists of a P2P-based pairwise registration, and an ICP refinement registration that implements the DSM. A new cost function, which utilizes simultaneously acquired forward direction 3D distance data and DSM information, is also introduced when proposing ICP. The excellence of the proposed position estimation method is evaluated through quantitative analysis of several experiment results.



17:45~18:00	WC08-6

Numerical singularity analysis of a tire model for real-time Traversability Simulation
Samuel Jung, Wan-Suk Yoo(Pusan national university, Korea, Republic of )

The real-time traversability simulation demands highly efficient and stable numerical model, but the longitudinal slip sometimes causes a numerical singularity problem from its definition. This problem makes it hard to integrate ODEs when using not only explicit integration method but also implicit integration method. This paper shows the reason for this problem on numerical point of view, and an improvement idea is proposed.