Time-Series JEPA for Predictive Remote Control under Capacity-Limited Networks

TL;DR

This paper introduces TS-JEPA, a novel framework that encodes sensory data into low-dimensional embeddings for predictive remote control, significantly improving communication efficiency and control performance in capacity-limited networks.

Contribution

The paper proposes TS-JEPA, a semantic-driven predictive control architecture that reduces data transmission and enhances control under limited network capacity through embedding and channel-aware scheduling.

Findings

Outperforms traditional control methods in communication efficiency

Reduces control cost in simulated inverted cart-pole systems

Enhances robustness and scalability in limited network environments

Abstract

In remote control systems, transmitting large data volumes (e.g., images, video frames) from wireless sensors to remote controllers is challenging when uplink capacity is limited (e.g., RedCap devices or massive wireless sensor networks). Furthermore, controllers often need only information-rich representations of the original data. To address this, we propose a semantic-driven predictive control combined with a channel-aware scheduling to enhance control performance for multiple devices under limited network capacity. At its core, the proposed framework, coined Time-Series Joint Embedding Predictive Architecture (TS-JEPA), encodes high-dimensional sensory data into low-dimensional semantic embeddings at the sensor, reducing communication overhead. Furthermore, TS-JEPA enables predictive inference by predicting future embeddings from current ones and predicted commands, which are directly used by a semantic actor model to compute control commands within the embedding space, eliminating the need to reconstruct raw data. To further enhance reliability and communication efficiency, a channel-aware scheduling is integrated to dynamically prioritize device transmissions based on channel conditions and age of information (AoI). Simulations on inverted cart-pole systems show that the proposed framework significantly outperforms conventional control baselines in communication efficiency, control cost, and predictive accuracy. It enables robust and scalable control under limited network capacity compared to traditional scheduling schemes.