Calibrate sibernetic worm movements to signals from neural data

[nmsutton]

With Sibernetic_NEURON by @skhayrulin et al. now enabling direct neural signal input into Sibernetic’s worm model movement there needs to be a translation of the neural signals into values that create appropriately scaled movement actions on the worm. The muscle_model project by @pgleeson, @vellamike et al. provides good work toward specifying how neural signals translate into muscle movement.

@nmsutton has done work toward implementing similar neural signal generation to that shown in Fig. 4C of a muscle_model related work article. That signaling can be used to direct bending motion of the worm using equation 5 and other formulas. Current programming work on the calibration module can be found here. Once signal to movement formulas are implemented they can input Sibernetic_NEURON signals and output scaled values for movement in Sibernetic model muscles that matches amounts specified in the article. @nmsutton is currently working on these efforts. Related to: #217, #11, #106, #107, #31

[nmsutton]

Further progress has been made on calibrating the neural signals to worm movement in Sibernetic. Using psuedo neural signaling values flipping muscle movements a crawling motion is achieved. The movements are based on the earlier described formula and in a pattern as seen in video. Examples showing my downsampled worm in a crawling motion from the muscle movements are: side view video and above view video.

Muscle objects are shown in red or blue. To enable the directed muscle movement the source code for sibernetic-NEURON has been expanded and changed in Sibernetic, the work in progress fork is here. Related to: #112, #106