Federated learning improves privacy and efficiency for 6g robot intelligence
Modality-Decoupled Federated Learning for Privacy-Preserving Embodied Intelligence in 6G
Artificial Intelligence
Summary
Connecting many robots over future 6G wireless networks lets them work together smarter, but sharing their knowledge can risk privacy and waste network resources. The authors propose FedMVLA, a new way for robots to learn together while keeping different types of information—like vision, language, and actions—separate and handled according to their needs. Their approach improves privacy protection, reduces the amount of data sent over the network, and speeds up learning. Tests show that FedMVLA helps robot teams complete tasks more successfully and efficiently, even when many devices are connected and when signals are disrupted.
What this means in practice
- •For industrial robotics teams: Coordinate many robots to learn complex manipulation tasks collaboratively while preserving data privacy and reducing network load over 6G wireless.
- •For wireless network engineers: Design 6G network slices optimized for latency-sensitive robot control data to improve reliability and speed in distributed AI systems.
Authors
Zhuodong Liu, Xiangyu Li, Chunhong Yuan, Hongyang Du, Bodong Shang, Qingqing Wu, Tony Q. S. Quek, Mohsen Guizani
Abstract
Sixth-generation (6G) wireless networks are expected to provide a key infrastructure for large-scale embodied intelligence, where heterogeneous robots collaborate through low-latency connectivity, edge intelligence, and distributed sensing. Vision-language-action (VLA) models offer a foundation by integrating visual perception, language understanding, and action generation into a unified closed-loop policy. However, training and adapting VLA models to distributed robotic agents introduce challenges in privacy protection, communication efficiency, and model heterogeneity. Existing federated learning (FL) methods overlook the intrinsic differences among vision, language, and action pathways in parameter scale, privacy exposure, update dynamics, and tolerance to compression or perturbation. To address this issue, this article proposes FedMVLA, a modality-decoupled FL framework for privacy-preserving embodied intelligence in 6G networks. FedMVLA incorporates three mechanisms: modality-aware federated aggregation (MAFA), modality-aware privacy allocation (MAPA), and modality-aware communication compression (MACO), together with a modality-sliced transport design that routes the precision-critical action stream through a protected ultra-reliable low-latency slice. A case study on federated robotic manipulation over the Third Generation Partnership Project (3GPP)-based wireless substrate, covering fading, co-channel interference, and malicious jamming, shows that FedMVLA achieves an 84.8% task success rate, exceeds FedAvg by 22.2 percentage points, sustains a widening margin when scaling to 128 clients across eight cells, and reduces the schedule-averaged per-client uplink model-update payload by 95.6% (approximately 96%), while keeping the 95th percentile (p95) of the round-critical uplink completion time near 1.5s.