Precision bin-picking using a 3D sensor and a 1D laser sensor

• Autor(a) principal: Mota, Joana Beatriz Carvalho
• Data de Publicação: 2018
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10773/27012
• topic: Bin-picking; Fanuc LR Mate 200iD; Kinect Sensor; Laser Sensor; ROS; ROSIndustrial; MoveIt; Point Cloud; PCL; Precision

The technique that is being used by a robot to grab objects that are randomly placed inside a box or on a pallet is called bin-picking. This process is of great interest in an industrial environment as it provides enhanced automation, increased production and cost reduction. Bin-picking has evolved greatly over the years due to tremendous strides empowered by advanced vision technology, software, and gripping solutions which are in constant development. However, the creation of a versatile system, capable of collecting any type of object without deforming it, regardless of the disordered environment around it, remains a challenge. To this goal, the use of 3D perception is unavoidable. Still, the information acquired by some lower cost 3D sensors is not very precise; therefore, the combination of this information with the one of other devices is an approach already in study. The main goal of this work is to develop a solution for the execution of a precise bin-picking process capable of grasping small and fragile objects without breaking or deforming them. This may be done by combining the information provided by two sensors: one 3D sensor (Kinect) used to analyse the workspace and identify the object, and a 1D laser sensor to determine the exact distance to the object when approaching it. Additionally, the developed system may be placed at the end of a manipulator in order to become an active perception unit. Once the global system of sensors, their controllers and the robotic manipulator are integrated into a ROS (Robot Operating System) infrastructure, the data provided by the sensors can be analysed and combined to provide a bin-picking solution. Finally, the testing phase demonstrated the viability and the reliability of the developed bin-picking process.