지능로봇을 위한 행위기반의 하이브리드 제어구조에 관한 연구 = (A) Study on behavior-based hybrid control architecture for intelligent robot

다국어 초록 (Multilingual Abstract)

In order to execute the various and complex tasks for a intelligent robot, more improvement is needed not only in mechanical system architecture but also in control system architecture.
This paper addresses behavior-based hybrid control architecture and a methodology of task representation to accomplish various and complex tasks successfully.
The suggested architecture consists of three layers such as deliberative, sequencing, and reactive layer. The deliberative layer has the function of interfacing with a user and executing task planning. The sequencing layer is classified into two groups. The first group is the controlling part that executes the internal process by managing the components in the reactive layer. The second group is the information part that extracts highly advanced information from raw data using various kinds of sophisticated and time-consuming algorithms. The reactive layer controls robot motion by executing relatively simple computation in real-time.
The suggested methodology of task representation is a multi-layer behavior model. Task is a collection of behavior and each behavior has sub-behaviors. Transitions between behaviors are defined by events and guard conditions. A process is a sequencing order of behaviors expressed in multi-layer model including error recovery logic. Simple and complex processes are stored and fetched in the task knowledge database for fast execution.
Finally, the suggested behavior-based hybrid control architecture and methodology of task representation are verified to be successful by 3D simulation.

목차 (Table of Contents)

• 목차
• 제 1 장 서론 = 1
• 1.1 연구 배경 및 목적 = 1
• 1.2 연구 내용 = 3
• 1.3 관련 연구 = 5
• 1.4 논문 구성 = 10
• 제 2 장 행위기반의 하이브리드 제어구조 = 11
• 2.1 행위 기반의 하이브리드 제어 = 11
• 2.2 Deliberative 층 = 15
• 2.3 Sequencing 층 = 18
• 2.4 Reactive 층 = 21
• 2.5 제어 모듈의 협력 다이어그램 = 23
• 제 3 장 태스크 표현 방법 = 26
• 3.1 다층 행위 모델 = 26
• 3.2 물건 나르기 심부름에 대한 상태-챠트 다이어그램 = 30
• 제 4 장 로봇 시뮬레이션 = 33
• 제 5 장 결론 = 44
• 참고 문헌 = 45
• ABSTRACT = 48