ADAPTIVE PICK AND PLACE APPROACH USING ROS FRAMEWORK

• Tipo: Artigo em Livro de Atas de Conferência Internacional
• Data: 2015

Autores

Tavares, P (Autor) Outra; Lima, J (Autor) Outra; Pedro Luís Cerqueira Gomes da Costa (Autor) FEUP; António Paulo Moreira (Autor) FEUP

Ata de Conferência Internacional

Título: PROCEEDINGS OF THE EUROPEAN CONFERENCE ON DATA MINING 2015 AND INTERNATIONAL CONFERENCES ON INTELLIGENT SYSTEMS AND AGENTS 2015 AND THEORY AND PRACTICE IN MODERN COMPUTING 2015

Páginas: 89-96

European Conference on Data Mining (ECDM) / International Conference on Intelligent Systems and Agents / International Conference on Theory and Practice in Modern Computing

Las Palmas, SPAIN, JUL 22-24, 2015

Indexação

ISI Web of Knowledge - 0 Citações

Scopus - 0 Citações

Outras Informações

ID Authenticus: P-00K-H2A

Abstract (EN): The field of Robotics has become one of the most rapidly growing fields in the research and technological world. The development of flexible robots represents the possibility of them becoming a highly efficient operator in the industrial environment. Intelligent robots present key characteristics that enable the streamlining of automated processes associated to industry. Adding the adaptive component to such robots facilitates the design of solutions for a wide range of problems. Pick and Place operations have attracted considerable interest from the research and industrial community as they present one of the most effective solutions to typical problems such as handling or transportation. Another aspect to consider when developing a robotic solution for pick and place approaches is the methodology for recognition of the objects to be handled. In this paper, it will be presented a methodology that can be applicable to different scenarios in order to both identify the objects of a given scene and reply to the need of handling those objects. Furthermore it will be presented one specific case study that used the proposed methodology, the Amazon Picking Challenge - a challenge aiming to develop solutions for the complete automation of a dispatching warehouse. Our proposed methodology was built using the Robotic Operative System (ROS) framework and is based in three tiers: recognition, movement / actuation and control. ROS allows the decomplexation of typical problems associated to robotics as this framework promotes the development of modular and simple software that together fulfill the state-of-art requests of the industry. Since ROS is becoming an important tool in robotics, using a methodology developed in ROS allows for the development of a standard approach to pick and place operations. Another advantage of our methodology is the ability to have a robot safely and efficiently inserted in an unknown environment. This is possible due to adaptive control tier. Proposed improvements to currently available methods will be also described in this project. Throughout the document, the importance of this project and the development of novel robots will be described taken into consideration the need for robots in the industrial setting.

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 8