ABSTRACT
The effectiveness of human-robot interactions critically depends on the success of computational efforts to emulate human inference of intent, anticipation of action, and coordination of movement. To this end, we developed two models that leverage a well described feature of human movement: Gaussian-shaped submovements in velocity profiles, to act as robotic surrogates for human inference and trajectory planning in a handover task. We evaluated both models based on how early in a handover movement the inference model can obtain accurate estimates of handover location and timing, and how similar model trajectories are to human receiver trajectories. Initial results using one participant dyad demonstrate that our inference model can accurately predict location and handover timing, while the trajectory planner can use these predictions to provide a human-like trajectory plan for the robot. This approach delivers promising performance while remaining grounded in physiologically meaningful Gaussian-shaped velocity profiles of human motion.
- Arash Ajoudani, Andrea Maria Zanchettin, Serena Ivaldi, Alin Albu-Schäffer, Kazuhiro Kosuge, and Oussama Khatib. 2018. Progress and prospects of the human–robot collaboration. Autonomous Robots 42, 5 (2018), 957–975.Google ScholarDigital Library
- Marco Costanzo, Giuseppe De Maria, and Ciro Natale. 2021. Handover Control for Human-Robot and Robot-Robot Collaboration. Frontiers in Robotics and AI 8 (2021), 132.Google ScholarCross Ref
- Auke Jan Ijspeert, Jun Nakanishi, and Stefan Schaal. 2002. Movement imitation with nonlinear dynamical systems in humanoid robots. In Proceedings 2002 IEEE International Conference on Robotics and Automation (Cat. No. 02CH37292), Vol. 2. IEEE, 1398–1403.Google ScholarCross Ref
- Hermano Igo Krebs, Mindy L. Aisen, Bruce T. Volpe, and Neville Hogan. 1999. Quantization of continuous arm movements in humans with brain injury. Proceedings of the National Academy of Sciences 96, 8 (1999), 4645–4649. https://doi.org/10.1073/pnas.96.8.4645 arXiv:https://www.pnas.org/content/96/8/4645.full.pdfGoogle ScholarCross Ref
- Muriel Lang, Satoshi Endo, Oliver Dunkley, and Sandra Hirche. 2017. Object handover prediction using gaussian processes clustered with trajectory classification. arXiv preprint arXiv:1707.02745(2017).Google Scholar
- Alberto Leardini, Fabio Biagi, Andrea Merlo, Claudio Belvedere, and Maria Grazia Benedetti. 2011. Multi-segment trunk kinematics during locomotion and elementary exercises. Clinical Biomechanics 26, 6 (2011), 562–571.Google ScholarCross Ref
- Ester Martinez-Martin and Angel P del Pobil. 2018. Personal robot assistants for elderly care: an overview. Personal assistants: Emerging computational technologies (2018), 77–91.Google Scholar
- Jan Peters and Stefan Schaal. 2008. Reinforcement learning of motor skills with policy gradients. Neural networks 21, 4 (2008), 682–697.Google Scholar
- Stefan Schaal. 2006. Dynamic Movement Primitives -A Framework for Motor Control in Humans and Humanoid Robotics. Springer Tokyo, Tokyo, 261–280. https://doi.org/10.1007/4-431-31381-8_23Google Scholar
- Kyle Strabala, Min Kyung Lee, Anca Dragan, Jodi Forlizzi, Siddhartha S Srinivasa, Maya Cakmak, and Vincenzo Micelli. 2013. Toward seamless human-robot handovers. Journal of Human-Robot Interaction 2, 1 (2013), 112–132.Google ScholarDigital Library
- Holger Urbanek, A Albu-Schaffer, and Patrick van der Smagt. 2004. Learning from demonstration: repetitive movements for autonomous service robotics. In 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)(IEEE Cat. No. 04CH37566), Vol. 4. IEEE, 3495–3500.Google ScholarCross Ref
- C. K. I. Williams and C. E. Rasmussen. 2006. Gaussian processes for machine learning. Vol. 2. MIT press Cambridge, MA.Google Scholar
- Daniel M Wolpert, Jörn Diedrichsen, and J Randall Flanagan. 2011. Principles of sensorimotor learning. Nature Reviews Neuroscience 12, 12 (2011), 739–751.Google ScholarCross Ref
- Wei Yang, Chris Paxton, Maya Cakmak, and Dieter Fox. 2020. Human grasp classification for reactive human-to-robot handovers. In 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 11123–11130.Google ScholarCross Ref
Recommendations
User Preferences in Occupational Sedentary Behaviour Digital Interventions: Design and Evaluation of a Low-Intrusive Software Tool
PETRA '22: Proceedings of the 15th International Conference on PErvasive Technologies Related to Assistive EnvironmentsOccupational sedentariness has been recognized as a public health issue as the workforce is increasingly employed in low activity jobs. Supporting the employees with digital tools that increase awareness of occupational sedentary behaviour can ...
Determining Best Hardware, Software and Data Structures for Worker Guidance during a Complex Assembly Task
PETRA '22: Proceedings of the 15th International Conference on PErvasive Technologies Related to Assistive EnvironmentsA widespread challenge in the industrial domain is the modernization and digitization of assembly processes involving human workers to increase production efficiency and thus stay competitive with rival companies. Specifically, in assembly processes ...
An Adaptive Tilting Interface to Alleviate Motion Sickness for Passengers in Vehicles
PETRA '22: Proceedings of the 15th International Conference on PErvasive Technologies Related to Assistive EnvironmentsWith today’s busy smartphone-based lifestyles and with the increase in the number of autonomous vehicles, there is a need for passengers to be able to perform tasks during their commute. However, motion sickness is elevated when passengers look at ...
Comments