Learning End-to-end Multimodal Sensor Policies for Autonomous Navigation

Guan-Horng Liu, Avinash Siravuru, Sai Prabhakar, Manuela Veloso, George Kantor
Proceedings of the 1st Annual Conference on Robot Learning, PMLR 78:249-261, 2017.

Abstract

We proposed a multimodal end-to-end policy based on deep reinforcement learning (DRL) that leverages sensor fusion to reduced performance drops in noisy environment from 50% to 10% compared with the baseline and makes the policy functional even in the face of partial sensor failure by using a novel stochastic technique called Sensor Dropout to reduce sensitivity to any sensor subset, and a new auxiliary loss on policy network along with standard DRL loss that reduces the action variations.

Cite this Paper

BibTeX
@InProceedings{pmlr-v78-liu17a, title = {Learning End-to-end Multimodal Sensor Policies for Autonomous Navigation}, author = {Liu, Guan-Horng and Siravuru, Avinash and Prabhakar, Sai and Veloso, Manuela and Kantor, George}, booktitle = {Proceedings of the 1st Annual Conference on Robot Learning}, pages = {249--261}, year = {2017}, editor = {Levine, Sergey and Vanhoucke, Vincent and Goldberg, Ken}, volume = {78}, series = {Proceedings of Machine Learning Research}, month = {13--15 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v78/liu17a/liu17a.pdf}, url = {https://proceedings.mlr.press/v78/liu17a.html}, abstract = {We proposed a multimodal end-to-end policy based on deep reinforcement learning (DRL) that leverages sensor fusion to reduced performance drops in noisy environment from 50% to 10% compared with the baseline and makes the policy functional even in the face of partial sensor failure by using a novel stochastic technique called Sensor Dropout to reduce sensitivity to any sensor subset, and a new auxiliary loss on policy network along with standard DRL loss that reduces the action variations.} }
Endnote
%0 Conference Paper %T Learning End-to-end Multimodal Sensor Policies for Autonomous Navigation %A Guan-Horng Liu %A Avinash Siravuru %A Sai Prabhakar %A Manuela Veloso %A George Kantor %B Proceedings of the 1st Annual Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2017 %E Sergey Levine %E Vincent Vanhoucke %E Ken Goldberg %F pmlr-v78-liu17a %I PMLR %P 249--261 %U https://proceedings.mlr.press/v78/liu17a.html %V 78 %X We proposed a multimodal end-to-end policy based on deep reinforcement learning (DRL) that leverages sensor fusion to reduced performance drops in noisy environment from 50% to 10% compared with the baseline and makes the policy functional even in the face of partial sensor failure by using a novel stochastic technique called Sensor Dropout to reduce sensitivity to any sensor subset, and a new auxiliary loss on policy network along with standard DRL loss that reduces the action variations.
APA
Liu, G., Siravuru, A., Prabhakar, S., Veloso, M. & Kantor, G.. (2017). Learning End-to-end Multimodal Sensor Policies for Autonomous Navigation. Proceedings of the 1st Annual Conference on Robot Learning, in Proceedings of Machine Learning Research 78:249-261 Available from https://proceedings.mlr.press/v78/liu17a.html.

Related Material