Augmenting sampling based controllers with machine learning

@inproceedings{Rajamki2017AugmentingSB,
  title={Augmenting sampling based controllers with machine learning},
  author={Joose Rajam{\"a}ki and Perttu H{\"a}m{\"a}l{\"a}inen},
  booktitle={Symposium on Computer Animation},
  year={2017}
}
Efficient learning of 3D character control still remains an open problem despite of the remarkable recent advances in the field. We propose a new algorithm that combines planning by a sampling-based model-predictive controller and learning from the planned control, which is very noisy. We combine two methods of learning: 1) immediate but imprecise nearest-neighbor learning, and 2) slower but more precise neural net-work learning. The nearest neighbor learning allows to rapidly latch on to new… CONTINUE READING

References

Publications referenced by this paper.
Showing 1-5 of 5 references

Similar Papers

Loading similar papers…