Article - Open Access.

Idioma principal | Segundo idioma

Haptic Learning Towards Neural-Network-based adaptive Cobot Path-Planning for unstructured spaces

Haptic Learning Towards Neural-Network-based adaptive Cobot Path-Planning for unstructured spaces

Rossi, Gabriella; Nicholas, Paul;

Article:

Collaborative Robots, or Cobots, bring new possibilities for human-machineinteraction within the fabrication process, allowing each actor to contribute withtheir specific capabilities. However creative interaction brings unexpectedchanges, obstacles, complexities and non-linearities which are encountered inreal time and cannot be predicted in advance. This paper presents anexperimental methodology for robotic path planning using Machine Learning.The focus of this methodology is obstacle avoidance. A neural network isdeployed, providing a relationship between the robot's pose and its surroundings,thus allowing for motion planning and obstacle avoidance, directly integratedwithin the design environment. The method is demonstrated through a series ofcase-studies. The method combines haptic teaching with machine learning tocreate a task specific dataset, giving the robot the ability to adapt to obstacleswithout being explicitly programmed at every instruction. This opens the door toshifting to robotic applications for construction in unstructured environments,where adapting to the singularities of the workspace, its occupants and activitiespresents an important computational hurdle today.

Article:

Collaborative Robots, or Cobots, bring new possibilities for human-machineinteraction within the fabrication process, allowing each actor to contribute withtheir specific capabilities. However creative interaction brings unexpectedchanges, obstacles, complexities and non-linearities which are encountered inreal time and cannot be predicted in advance. This paper presents anexperimental methodology for robotic path planning using Machine Learning.The focus of this methodology is obstacle avoidance. A neural network isdeployed, providing a relationship between the robot's pose and its surroundings,thus allowing for motion planning and obstacle avoidance, directly integratedwithin the design environment. The method is demonstrated through a series ofcase-studies. The method combines haptic teaching with machine learning tocreate a task specific dataset, giving the robot the ability to adapt to obstacleswithout being explicitly programmed at every instruction. This opens the door toshifting to robotic applications for construction in unstructured environments,where adapting to the singularities of the workspace, its occupants and activitiespresents an important computational hurdle today.

Palavras-chave: ,

Palavras-chave: ,

DOI: 10.5151/proceedings-ecaadesigradi2019_280

Referências bibliográficas
  • [1] .
Como citar:

Rossi, Gabriella; Nicholas, Paul; "Haptic Learning Towards Neural-Network-based adaptive Cobot Path-Planning for unstructured spaces", p. 201-210 . In: Proceedings of 37 eCAADe and XXIII SIGraDi Joint Conference, “Architecture in the Age of the 4Th Industrial Revolution”, Porto 2019, Sousa, José Pedro; Henriques, Gonçalo Castro; Xavier, João Pedro (eds.). São Paulo: Blucher, 2019.
ISSN 2318-6968, DOI 10.5151/proceedings-ecaadesigradi2019_280

últimos 30 dias | último ano | desde a publicação


downloads


visualizações


indexações