Publicação
Human-like motion generation through waypoints for collaborative robots in Industry 5.0
| Resumo: | Industry 4.0 has motivated the scientific community to innovate solutions to ensure that companies maintain levels of competitiveness and meet customer demands, which are increasingly higher due to mass customization [Villani et al. (2018)]. Robots need more flexibility, intuitive and user-friendly programming methods so that they can be easily reprogrammed for new tasks. Thus, collaborative robots have emerged, which are smaller, safer, and most importantly, are able to share the workspace with human operators [Villani et al. (2018)]. Moreover, there are already predictions regarding Industry 5.0, where humans and robots will coexist in their daily routines [Schaal (2007)]. This dissertation proposes a trajectory planning method that addresses the above needs. This method allows operators to easily program the robot for a new task by defining mandatory positions -waypoints- of the trajectory. Waypoints can be defined by physically manipulating the robot or by using the joystick built into the teach pendant robot. The generated trajectory is based on the minimum-jerk model introduced by Flash and Hogan (1985), which guarantees both quantitative and qualitative human characteristics. Such properties have a very positive impact on operators’ well-being and productivity [Koppenborg et al. (2017),El Zaatari et al. (2019)]. The proposed method is validated in a quality inspection scenario in an industry context. Specifically, the user defines waypoints, which correspond to the position of the eye angle to inspect the plates, and subsequently the robot manipulates them through the mandatory points in a human-like manner. The planner allows smooth, fluent, and intuitive movements through the waypoints. Although the resulting movements have human characteristics, we cannot absolutely claim that they are human movements, since there are no experiments on humans with waypoints. |
|---|---|
| Autores principais: | Pereira, João André Correia Queiroga |
| Assunto: | Waypoints Human-like Trajectory generation Optimal control Interior-point constraints Collaborative robots UR10 Industry 4.0 Industry 5.0 Geração de trajetórias Controlo ótimo Restrições de pontos interiores Robôs colaborativos Indústria 4.0 Indústria 5.0 |
| Ano: | 2021 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade do Minho |
| Idioma: | português |
| Origem: | RepositóriUM - Universidade do Minho |
| Resumo: | Industry 4.0 has motivated the scientific community to innovate solutions to ensure that companies maintain levels of competitiveness and meet customer demands, which are increasingly higher due to mass customization [Villani et al. (2018)]. Robots need more flexibility, intuitive and user-friendly programming methods so that they can be easily reprogrammed for new tasks. Thus, collaborative robots have emerged, which are smaller, safer, and most importantly, are able to share the workspace with human operators [Villani et al. (2018)]. Moreover, there are already predictions regarding Industry 5.0, where humans and robots will coexist in their daily routines [Schaal (2007)]. This dissertation proposes a trajectory planning method that addresses the above needs. This method allows operators to easily program the robot for a new task by defining mandatory positions -waypoints- of the trajectory. Waypoints can be defined by physically manipulating the robot or by using the joystick built into the teach pendant robot. The generated trajectory is based on the minimum-jerk model introduced by Flash and Hogan (1985), which guarantees both quantitative and qualitative human characteristics. Such properties have a very positive impact on operators’ well-being and productivity [Koppenborg et al. (2017),El Zaatari et al. (2019)]. The proposed method is validated in a quality inspection scenario in an industry context. Specifically, the user defines waypoints, which correspond to the position of the eye angle to inspect the plates, and subsequently the robot manipulates them through the mandatory points in a human-like manner. The planner allows smooth, fluent, and intuitive movements through the waypoints. Although the resulting movements have human characteristics, we cannot absolutely claim that they are human movements, since there are no experiments on humans with waypoints. |
|---|