The paper presents a case study on automating the process of blasting charger in underground mines using a robotic system. The authors describe the challenges of this industrial application, which include working in a previously unknown environment, requiring reliable computer vision algorithms, fast reaction to unforeseen events, and the need for operator supervision and adjustable autonomy.
To address these challenges, the authors propose a hybrid approach that combines a high-level FSM and lower-level BTs. The FSM is responsible for transitioning the execution between different operational modes, such as scanning the working area, detecting holes to charge, planning the charging sequence, and executing the charging process. Each of these operational modes is then implemented as a BT, which enables the robotic system to be reactive and robust to failures during the execution stage.
The authors highlight the benefits of this combined approach, including:
The paper also discusses the lessons learned from the implementation, such as the importance of finding the right balance between modularity and readability when designing the BTs, and the challenges of modeling cyclic behaviors in BTs.
The proposed solution was successfully demonstrated in a real underground mine, and the authors provide insights into the applicability of BTs in industrial contexts and the technical suggestions on BT design practices.
To Another Language
from source content
arxiv.org
Глибші Запити