Robot task planning is an important problem for autonomous robots in long-horizon challenging tasks. As large pre-trained models have demonstrated superior planning ability, recent research investigates utilizing large models to achieve autonomous planning for robots in diverse tasks. However, since the large models are pre-trained with Internet data and lack the knowledge of real task scenes, large models as planners may make unsafe decisions that hurt the robots and the surrounding environments. To solve this challenge, we propose a novel Safe Planner framework, which empowers safety awareness in large pre-trained models to accomplish safe and executable planning. In this framework, we develop a safety prediction module to guide the high-level large model planner, and this safety module trained in a simulator can be effectively transferred to real-world tasks. The proposed Safe Planner framework is evaluated on both simulated environments and real robots. The experiment results demonstrate that Safe Planner not only achieves state-of-the-art task success rates, but also substantially improves safety during task execution.

Some WLAN (Wireless Local Area Network) roaming systems such as eduroam may experience increased authentication latency or failures, especially when authentication requests traverse multiple proxy nodes to reach their home Identity Provider (IdP). Existing systems rely on static configurations and lack adaptability to path condition changes, often leading to degraded performance.This paper presents a distributed path selection mechanism based on passive observation of EAP-TLS authentication traffic. Each proxy node locally evaluates downstream paths using metrics such as success rate and Round-Trip Time (RTT), enabling decentralized and adaptive forwarding decisions without centralized coordination.Experiments on a multi-hop simulation platform demonstrate that the proposed system improves success rates, reduces latency, and quickly recovers from failures. This work highlights the feasibility of using real authentication feedback as a lightweight yet effective basis for path optimization in federated WLAN environments.