Closed-form control with spike coding networks
This work addresses the challenge of deploying fast and efficient task-specific on-chip spiking controllers with biologically realistic activity, offering a solution for low-power implementations, though it is incremental as it builds upon existing SCN theory.
The paper tackled the problem of efficient and robust control using spiking neural networks by extending Spike Coding Networks theory to incorporate closed-form optimal estimation and control, resulting in a spiking equivalent of a linear-quadratic-Gaussian controller that demonstrated robust control of simulated systems like spring-mass-damper and cart-pole under various perturbations without needing learning or optimization.
Efficient and robust control using spiking neural networks (SNNs) is still an open problem. Whilst behaviour of biological agents is produced through sparse and irregular spiking patterns, which provide both robust and efficient control, the activity patterns in most artificial spiking neural networks used for control are dense and regular -- resulting in potentially less efficient codes. Additionally, for most existing control solutions network training or optimization is necessary, even for fully identified systems, complicating their implementation in on-chip low-power solutions. The neuroscience theory of Spike Coding Networks (SCNs) offers a fully analytical solution for implementing dynamical systems in recurrent spiking neural networks -- while maintaining irregular, sparse, and robust spiking activity -- but it's not clear how to directly apply it to control problems. Here, we extend SCN theory by incorporating closed-form optimal estimation and control. The resulting networks work as a spiking equivalent of a linear-quadratic-Gaussian controller. We demonstrate robust spiking control of simulated spring-mass-damper and cart-pole systems, in the face of several perturbations, including input- and system-noise, system disturbances, and neural silencing. As our approach does not need learning or optimization, it offers opportunities for deploying fast and efficient task-specific on-chip spiking controllers with biologically realistic activity.