Graceful Degradation Under Noise on Brain Inspired Robot Controllers
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
BaxterArm_VREP_simulation_data/square File names adjusted to match the paper Sep 1, 2016
VREP_scenes File names adjusted to match the paper Sep 1, 2016
BEE_Simulator_ArmControl_VREP_LSM_DATA-GENERATOR.ipynb Source code used for the ICONIP2016 paper May 12, 2016
BEE_Simulator_ArmControl_VREP_LSM_DATA-TESTER-Modular-Robustness.ipynb File names adjusted to match the paper Sep 1, 2016
BEE_Simulator_ArmControl_VREP_LSM_DATA-TESTER-Monolithic-Robustness.ipynb File names adjusted to match the paper Sep 1, 2016
BEE_Simulator_ArmControl_VREP_LSM_DATA-TESTER-VREP.ipynb File names adjusted to match the paper Sep 1, 2016
BEE_Simulator_ArmControl_VREP_LSM_LINEAR_REGRESSION-Modular.ipynb File names adjusted to match the paper Sep 1, 2016
BEE_Simulator_ArmControl_VREP_LSM_LINEAR_REGRESSION-Monolithic.ipynb File names adjusted to match the paper Sep 1, 2016
BEE_Simulator_ArmControl_VREP_trajectories-generator_v1-SQUARE.ipynb File names adjusted to match the paper Sep 1, 2016
Dynamic_Time_Warping.ipynb Source code used for the ICONIP2016 paper May 12, 2016
ICONIP2016_figures_generator.ipynb File names adjusted to match the paper Sep 1, 2016
ICONIP2016_preprint.pdf Preprint version of the paper Nov 3, 2016
LICENSE Create LICENSE Jul 25, 2017
README.md Update README.md Jul 25, 2017
dtw_C.pyc Source code used for the ICONIP2016 paper May 12, 2016
dtw_python.so Source code used for the ICONIP2016 paper May 12, 2016
membrane_lowpass_md.pyc Source code used for the ICONIP2016 paper May 12, 2016
remoteApi.dylib Source code used for the ICONIP2016 paper May 12, 2016
save_load_file.pyc Source code used for the ICONIP2016 paper May 12, 2016
vrep.py Source code used for the ICONIP2016 paper May 12, 2016
vrepConst.py Source code used for the ICONIP2016 paper May 12, 2016

README.md

Experiments used for the paper submitted to presented at ICONIP2016

Graceful Degradation under Noise on Brain Inspired Robot Controllers

Abstract

How can we build robot controllers that are able to work under harsh conditions, but without experiencing catastrophic failures? As seen on the recent Fukushima's nuclear disaster, standard robots break down when exposed to high radiation environments. Here we present the results from two arrangements of Spiking Neural Networks, based on the Liquid State Machine (LSM) framework, that were able to gracefully degrade under the effects of a noisy current injected directly into each simulated neuron. These noisy currents could be seen, in a simplified way, as the consequences of exposition to non-destructive radiation. The results show that not only can the systems withstand noise, but one of the configurations, the Modular Parallel LSM, actually improved its results, in a certain range, when the noise levels were increased. Also, the robot controllers implemented in this work are suitable to run on a modern, power efficient neuromorphic hardware such as SpiNNaker.

(this work is based on github.com/ricardodeazambuja/IJCNN2016)

  1. The trajectories are generated using a simulated BAXTER robot inside V-REP:

    • BEE_Simulator_ArmControl_VREP_trajectories-generator_v1-SQUARE.ipynb
    • /VREP_scenes/Baxter_IK_felt_pen_pick-and-place_learning_ICONIP2016.ttt
  2. Training data (output spikes) are generated using the notebook:

    • BEE_Simulator_ArmControl_VREP_LSM_DATA-GENERATOR.ipynb
  3. After the generation of the training data, it is necessary to train the readouts. This is done by:

    • BEE_Simulator_ArmControl_VREP_LSM_LINEAR_REGRESSION-Parallel_Maass.ipynb
    • BEE_Simulator_ArmControl_VREP_LSM_LINEAR_REGRESSION.ipynb
  4. With all the readout weights defined, it's possible to verify the system using only the LSMs:

    • BEE_Simulator_ArmControl_VREP_LSM_DATA-TESTER-VREP.ipynb

OBS:

BEE SNN simulator:

https://github.com/ricardodeazambuja/BEE

Dynamic Time Warping (DTW):

https://github.com/ricardodeazambuja/DTW

Python scripts in general:

https://github.com/ricardodeazambuja/Python-UTILS

V-REP simulator:

http://www.coppeliarobotics.com/downloads.html

Preprint version:

https://github.com/ricardodeazambuja/ICONIP2016/blob/master/ICONIP2016_preprint.pdf

Bibtex citation:

https://github.com/ricardodeazambuja/ricardodeazambuja.github.io/raw/master/public/citations/azambuja_graceful_2016.bib

Final Springer Link version:

http://link.springer.com/chapter/10.1007/978-3-319-46687-3_21

Related works:

https://github.com/ricardodeazambuja/IJCNN2016
https://github.com/ricardodeazambuja/IJCNN2017
https://github.com/ricardodeazambuja/IJCNN2017-2
https://github.com/ricardodeazambuja/I2MTC2017-LSMFusion