Found 3 results.

discovery of complex behaviors through contact-invariant optimization

Author(s): Igor Mordatch, Emanuel Todorov, Zoran Popovic
Venue: ACM Transactions on Graphics
Year Published: 2012
Keywords: planning, contact dynamics, trajectory optimization, locomotion, reinforcement learning
Expert Opinion: The paper demonstrates that with an accurate internal model, planning of complex behaviors including contacts and dynamic interaction with the environment is possible from scratch. I see it as an important result supporting the need for good internal representations, which in the case of real-world interactions need to be at least partially learned.

assessing grasp stability based on learning and haptic data

Author(s): Yasemin Bekiroglu, Janne Laaksonen, Jimmy Alison Jurgensen, Ville Kyrki and Danica Kragic
Venue: IEEE Transactions on Robotics
Year Published: 2011
Keywords: manipulation, visual perception, contact dynamics
Expert Opinion: "Learning to grasp" can actually imply a lot of different learning problems. We often think about grasp synthesis, i.e., the problem of determining where to place the hand to achieve a stable grasp (I strongly recommend reading Data-Driven Grasp Synthesis- a Survey for more on this topic). This paper focuses on the important problem of using multiple sensor modalities to determine if an executed grasp attempt resulted in a stable grasp. As robot learning researchers, it is important to consider how problems can be approached from different directions and how different information sources can be incorporated and change the problem. One should also think about robustness and consider how learning factors into monitoring skill executions for errors.

mujoco (software)

Author(s): Emanuel Todorov, Tom Erez and Yuval Tassa
Venue: Software
Year Published: 2012
Keywords: contact dynamics, dynamical systems, reinforcement learning
Expert Opinion: Mujoco (together with Bullet) is probably the most popular simulator for learning based robotics. It's built with robotics research in mind (as opposed to many other simulators that were initially built for games), and consequently has a lot of flexibility around things that matter for robotics, like physical realism, sensors, actuation models, tendons, and so on. It has a clean and easy-to-use interface and is extremely fast, both which are useful features for iterating on new research.