SDP: Spiking Diffusion Policy for Robotic Manipulation with Learnable Channel-Wise Membrane Thresholds

TL;DR

This paper presents a novel Spiking Diffusion Policy (SDP) that integrates spiking neural networks with learnable thresholds and diffusion models, achieving high performance and energy efficiency in robotic manipulation tasks.

Contribution

The paper introduces a new SDP model with learnable channel-wise membrane thresholds and a U-Net based SNN architecture, improving efficiency and performance in manipulation tasks.

Findings

Achieves performance comparable to ANN models.

Faster convergence than baseline SNN methods.

Reduces dynamic energy consumption by 94.3%.

Abstract

This paper introduces a Spiking Diffusion Policy (SDP) learning method for robotic manipulation by integrating Spiking Neurons and Learnable Channel-wise Membrane Thresholds (LCMT) into the diffusion policy model, thereby enhancing computational efficiency and achieving high performance in evaluated tasks. Specifically, the proposed SDP model employs the U-Net architecture as the backbone for diffusion learning within the Spiking Neural Network (SNN). It strategically places residual connections between the spike convolution operations and the Leaky Integrate-and-Fire (LIF) nodes, thereby preventing disruptions to the spiking states. Additionally, we introduce a temporal encoding block and a temporal decoding block to transform static and dynamic data with timestep $T_S$ into each other, enabling the transmission of data within the SNN in spike format. Furthermore, we propose LCMT to enable the adaptive acquisition of membrane potential thresholds, thereby matching the conditions of varying membrane potentials and firing rates across channels and avoiding the cumbersome process of manually setting and tuning hyperparameters. Evaluating the SDP model on seven distinct tasks with SNN timestep $T_S=4$, we achieve results comparable to those of the ANN counterparts, along with faster convergence speeds than the baseline SNN method. This improvement is accompanied by a reduction of 94.3\% in dynamic energy consumption estimated on 45nm hardware.