The project SENSOPAC: Sensorimotor structuring of Perception and Action is funded under the EU Framework 6 IST Cognitive Systems initiative for a period of 48 months. The project involves a total of 11 partners across Europe including research groups from Cambridge (D. Wolpert), DLR Germany (P. van der Smagt), SONY France (O. Coenen), Erasmus MC (C. De Zeeuw), Pavia (E. D'Angelo), UMEA-Sweden (R.Johansson), LUND-Sweden (Jorntell, Ekerot), Bar-Ilan (Cohen), UGR-Spain (Ros), Altjira (Arnold)  besides Edinburgh
The overall aims of the SENSOPAC project are to: (i) Understand the sensorimotor foundation of perception and cognition, (ii) Improve our understanding of the neurobiological substrate for action-perception systems, (iii) Build a physical system for haptic cognition, (iv) Use probabilistic techniques to investigate cognition in the brain, (v) Develop real-time neuromorphic and computing platforms for cognitive robotics. The project will deal with multiple sensory modalities like haptic, proprioceptive, position and force feedback.

The Edinburgh group will be responsible for the machine learning approaches to discovering sensorimotor contingencies from high dimensional sensorimotor data and development of novel control methods for use with  redundant actuation and multiple, varying contexts. We will work with both artificial and real world (robot and human) data.

Datasets and software

Some of our datasets, software packages, and the code we use to run our experiments may be of interest to other SENSOPAC partners. During the course of the project, we make relevant material available on this webpage.

SENSOPAC related publication output:

• Georgios Petkos and Sethu Vijayakumar,
  Load estimation and control using learned dynamics models,
  Proc. IEEE International Conference on Intelligent Robots and Systems (IROS), San Diego, USA (2007).
• Adrian Haith and Sethu Vijayakumar,
  Robustness of VOR and OKR adaptation under kinematics and dynamics transformations,
  Proc. Sixth IEEE international Conference on Development and Learning (ICDL '07), London,(2007). [pdf]
• Heiko Hoffmann, Georgios Petkos, Sebastian Bitzer and Sethu Vijayakumar,
  Sensor Assisted Adaptive Motor Control under continuously varying Context,
  Proc. International Conference on Informatics in Control, Automation and Robotics (ICINCO '07), Angers, France(2007). [pdf]
• Georgios Petkos and Sethu Vijayakumar,
  Context Estimation and Learning Control through Latent Variable Extraction: From discrete to continuous contexts, Proc. IEEE International Conference on Robotics and Automation (ICRA '07), Rome, Italy (2007). [pdf]
• Masashi Sugiyama, Hirotaka Hachiya, Christopher Towell and Sethu Vijayakumar,
  Value Function Approximation on Non-Linear Manifolds for Robot Motor Control,
  Proc. IEEE International Conference on Robotics and Automation (ICRA '07), Rome, Italy (2007). [pdf]
• Narayanan Edakunni, Stefan Schaal and Sethu Vijayakumar,
  Kernel Carpentry for Online Regression using Randomly Varying Coefficient Model,
  Proc. International Joint Conference on Artificial Intelligence (IJCAI '07), Hyderabad, India (2007). [pdf]
• Timothy Hospedales, Joel Cartwright and Sethu Vijayakumar,
  Structure Inference for Bayesian Multisensory Perception and Tracking,
  Proc. International Joint Conference on Artificial Intelligence (IJCAI '07), Hyderabad, India (2007). [pdf]
• Graham McNeill and Sethu Vijayakumar,
  Part-based Probabilistic Point Matching Using Equivalence Constraints,
  Proc. Advances in Neural Information Processing Systems (NIPS '06), Vancouver, Canada (2006). [pdf]
• Matthew Howard , Michael Gienger, Christian Goerick and Sethu Vijayakumar,
  Learning Utility Surfaces for Movement Selection,
  Proc. IEEE International Conference on Robotics and Biomimetics (ROBIO), Kunmin, China (2006). [pdf]
• Georgios Petkos, Marc Toussaint and Sethu Vijayakumar,
  Learning Multiple Models of Non-Linear Dynamics for Control under Varying Contexts,
  Proc. International Conference on Artificial Neural Networks (ICANN '06), Athens, Greece (2006). [pdf]