Planning and acting framework under robot operating system

2nd International Conference on Engineering Sciences

Al Moadhen, A.A.^1 ; Abdulhussein, A.M.^2 ; Kamil, H.G.^1

College of Engineering, University of Kerbala, Kerbala, Iraq^1

University of Kerbala, Kerbala, Iraq^2

关键词: Additional flexibilities;    Autonomous Mobile Robot;    Autonomous Vehicles;    Environment modeling;    Robot operating system;    Robot operating systems (ROS);    Sequence of actions;    Task specifications;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/433/1/012090/pdf
DOI  :  10.1088/1757-899X/433/1/012090

来源: IOP

【 摘 要 】

A well-known Operating System (OS) in the field of robotics is the Robot Operating System (ROS), which provides a set of tools and software libraries that are used to build many modern robotics systems. A planning system is concerned with generating a sequence of actions (plan) that can then be used by the robotic system to achieve its tasks. However, autonomous systems cannot depend only on planning systems to accomplish such tasks. In order to support such systems, it is necessary to develop an acting system, whose role is to complement the planning system by endowing the robotic system with additional flexibility to accomplish its tasks. In this paper, a state-of-the-art framework, RosPlanAct, is described. This architecture is used for embedding task planning and acting systems into the ROS. The planning section uses the environment model (planning domain) and the task specifications (planning problem) to generate an appropriate sequence of actions (plan) for any robot tasks. Then, the acting part of the system receives this plan and acts upon its schedule to achieve the specified task. Throughout this paper, there are three main issues that the actor must cope with to avoid failure. These issues are plan refinement, re-planning, and reacting to exogenous events. The specification of the RosPlanAct architecture and a case study in autonomous mobile robot are explained. The case study involves an autonomous vehicle in scenarios that clearly demonstrate the flexibility, stability, and robustness of the presented approach.