Konstantinos Chatzilygeroudis
Konstantinos Chatzilygeroudis
We are simulating some simple robots (hexapod) in very simple environments (just floor) and using code from your examples (spd controller). Everything works very well until the moment we enable...
Hello, Great project! Thanks for sharing! I want to ask you whether you have made some comparisons with other libraries. For example: - Physics quality and speed compared to Bullet...
This PR adds GP gradient computations. ~~At the moment we output the gradient with respect to the kernel function and the input (i.e., the mean function is considered a static...
Lot of people state that Appveyor is better than travis and now they have [linux support](https://www.appveyor.com/docs/getting-started-with-appveyor-for-linux/).. Maybe at some point we want to change to Appveyor..
It'd be a good idea to use [pipelines (build stages)](https://docs.travis-ci.com/user/build-stages) of travis to install NLOpt, libcmaes and dependencies only once (and then use them if needed) instead of rebuilding them...
It seems MKL has an implementation for cholesky decomposition.. https://software.intel.com/en-us/node/521465 http://stackoverflow.com/questions/31231428/cholesky-with-scalapack This could be a nice option to Limbo since we spend a lot time there (esp. with likelihood optimization).
We should at some point change the API to allow vectorization of kernel functions. For example, we would like to do: `_kernel = K(X,X)` where `X` is a vector of...
http://jmlr.csail.mit.edu/papers/volume13/hennig12a/hennig12a.pdf
As discussed in #40 , we need to decide on what stopping criteria we should impose to the optimizers. A clean way would be for all of them to have...
Thanks for the nice simulator. Are there any plans for supporting mimic joints (that URDF supports)? We really need this feature. Thanks a lot! If needed I can provide test...