Blog

News and opinion


Back to overview

26.03.2020

Simulation of generic ANS Level 1 (=insects) with OP-L1 beta release

Our company is testing at the moment the beta release of our
operating system for complexity level 1, which is actually
the equivalent for the brains of insects, e.g. honey bees.
The capacity for this brain is max. 1 M neuronal cells.

For this purpose we are running different simulations for generic ANS,
which on their side are customized to a certain given body framework,
with configurations having various sensors and actors,
as well as the required internal organs to be managed.

The self-learning mechanism of our ANS within the digital brain
is the same as the biological counterpart, with same results.
Higher brain functions, which are known for birds, mammals
or even human are not possible with this level of complexity.

As a reminder, different levels of complexity for an ANS
require a corresponding operating system, L1 stand for insects.
For the next level of complexity, e.g. mammals, we will have the OP-L2,
while for the level 3, equivalent for humans, we will develop the OP-L3.

However, the ANS level 1 is the basis for all our operating systems,
as each further level of complexity integrates the previous one and
extends accordingly the brain capacity, as OP-L2 is an extension of OP-L1,
while the capacity is 1000 times higher than OP-L1, with max. 1 G neurons.

For OP-L1 and OP-L2 the FPGA-SiP is still enough to manage all ressources,
however for the OP-L3, means human, the capacity is much higher, therefore
we will replace the FPGA-SiP with 3DSOC from Skywater technology, which
will enable a digital brain made only with 3DSOC, using stackable boards.

We will start developing a real ANS L1 with an appropriate body,
immediately after we have tested and finished our OP-L1 beta release.
We will then implement the ANS L1 within our OP-L1 and the test
the entire embedded system consisting of body hw with OP-L1 and ANS-L1.

Even if we are developing a biological inspired ANS,
certain body parts can not be implemented digitally,
such as smell, taste and muscles, let alone the size and amount
of individual sensing/acting points of a biological organism.

Therefore, we are using digital sensors which are available,
same with servomodules with motors instead of muscle groups.
Nevertheless, the principle of connecting and processing
the corresponding sensors and actors is the same for each brain.



Back to overview