# Spiking Neural Networks for Continuous Control via End-to-End Model-Based Learning

arXiv: https://arxiv.org/abs/2509.05356 · HTML: https://mlsanity.com/paper/2509.05356/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 60 references that are in the map.
- First posted on arXiv 3 Sep 2025.

## Builds on (most cited first, 25 of 25)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Recurrent World Models Facilitate Policy Evolution](https://mlsanity.com/paper/1809.01999/) (2018)
- [Surrogate Gradient Learning in Spiking Neural Networks](https://mlsanity.com/paper/1901.09948/) (2019)
- [Orbit: A Unified Simulation Framework for Interactive Robot Learning Environments](https://mlsanity.com/paper/2301.04195/) (2023)
- [Deep Learning in Spiking Neural Networks](https://mlsanity.com/paper/1804.08150/) (2018)
- [Training Spiking Neural Networks Using Lessons From Deep Learning](https://mlsanity.com/paper/2109.12894/) (2021)
- [SuperSpike: Supervised learning in multi-layer spiking neural networks](https://mlsanity.com/paper/1705.11146/) (2017)
- [Accurate and efficient time-domain classification with adaptive spiking recurrent neural networks](https://mlsanity.com/paper/2103.12593/) (2021)
- [Active Inference in Robotics and Artificial Agents: Survey and Challenges](https://mlsanity.com/paper/2112.01871/) (2021)
- [Deep Reinforcement Learning with Population-Coded Spiking Neural Network for Continuous Control](https://mlsanity.com/paper/2010.09635/) (2020)
- [World Models and Predictive Coding for Cognitive and Developmental Robotics: Frontiers and Challenges](https://mlsanity.com/paper/2301.05832/) (2023)
- [Accurate online training of dynamical spiking neural networks through Forward Propagation Through Time](https://mlsanity.com/paper/2112.11231/) (2021)
- [Fully neuromorphic vision and control for autonomous drone flight](https://mlsanity.com/paper/2303.08778/) (2023)
- [Fluctuation-driven initialization for spiking neural network training](https://mlsanity.com/paper/2206.10226/) (2022)
- [Autonomous Driving with Spiking Neural Networks](https://mlsanity.com/paper/2405.19687/) (2024)
- [Fully Spiking Neural Network for Legged Robots](https://mlsanity.com/paper/2310.05022/) (2023)
- [Closed-loop spiking control on a neuromorphic processor implemented on the iCub](https://mlsanity.com/paper/2009.09081/) (2020)
- [Closed-form control with spike coding networks](https://mlsanity.com/paper/2212.12887/) (2022)
- [Many-Joint Robot Arm Control with Recurrent Spiking Neural Networks](https://mlsanity.com/paper/2104.04064/) (2021)
- [Learning Policies for Continuous Control via Transition Models](https://mlsanity.com/paper/2209.08033/) (2022)
- [Synaptic motor adaptation: A three-factor learning rule for adaptive robotic control in spiking neural networks](https://mlsanity.com/paper/2306.01906/) (2023)
- [Noisy Spiking Actor Network for Exploration](https://mlsanity.com/paper/2403.04162/) (2024)
- [Training Deep Spiking Auto-encoders without Bursting or Dying Neurons through Regularization](https://mlsanity.com/paper/2109.11045/) (2021)
- [Bioinspired Smooth Neuromorphic Control for Robotic Arms](https://mlsanity.com/paper/2209.02787/) (2022)
- [Spiking neurons as predictive controllers of linear systems](https://mlsanity.com/paper/2507.16495/) (2025)

## Abstract (arXiv)

Despite recent progress in training spiking neural networks (SNNs) for classification, their application to continuous motor control remains limited. Here, we demonstrate that fully spiking architectures can be trained end-to-end to control robotic arms with multiple degrees of freedom in continuous environments. Our predictive-control framework combines Leaky Integrate-and-Fire dynamics with surrogate gradients, jointly optimizing a forward model for dynamics prediction and a policy network for goal-directed action. We evaluate this approach on both a planar 2D reaching task and a simulated 6-DOF Franka Emika Panda robot with torque control. In direct comparison to non-spiking recurrent baselines trained under the same predictive-control pipeline, the proposed SNN achieves comparable task performance while using substantially fewer parameters. An extensive ablation study highlights the role of initialization, learnable time constants, adaptive thresholds, and latent-space compression as key contributors to stable training and effective control. Together, these findings establish spiking neural networks as a viable and scalable substrate for high-dimensional continuous control, while emphasizing the importance of principled architectural and training design.