Human Aware Cooperation in Industrial Manipulators

• Autor(a) principal: Furtado, Gonçalo Velho Cabral Pacheco
• Data de Publicação: 2023
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Texto Completo: http://hdl.handle.net/10362/181012
• topic: Hierarchical Architecture; Awareness; Cooperation; Safety; Manipulation; Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática

Industrial manipulators have grown to become the most adopted technology in the industry to help mitigate manual labor and shift the focus to the more core problems of the assembly line, but their adoption has created a physical and psychological barrier between them and the human workers. Therefore, this dissertation proposes a cognitive system that extends the capabilities of traditional autonomous systems by incorporating human-like abilities such as action recognition and cooperation. To achieve this, a hierarchical architecture is adopted, consisting of three phases: Awareness, Cooperation, and Safety. The former extracts knowledge from visual data captured from the physical world, while the Cooperation phase utilizes this knowledge to make logical decisions regarding suitable cooperative goals. Finally, the Security phase plans collision-free trajectories for the manipulator to execute and achieve the desired goals. Each phase consists of multiple modules, each addressing specific functionality requirements. This approach was implemented in the ROS environment where each module aggre- gates a set of nodes, some built to comply with the specific requirements others extracted from already advanced works. Its feasibility was tested in an experimental setup simulat- ing an industrial environment which comprised of a UR3e cobot from Universal Robots mounted with a two-finger 140mm gripper from Robotiq and an Intel RealSense D435i camera. The results demonstrated the effectiveness of the proposed architecture in coordinating cooperative responses, achieving collision-free trajectories, and promoting a cooperative environment between humans and industrial manipulators. This research contributes to the field of human-robot cooperation and provides practical insights for engineers and industry professionals. By leveraging the potential of the hierarchical architecture and implementing the recommendations outlined in this research, industries can play a crucial role in achieving a more efficient and cooperative future between humans and machines.