arXiv subjects

Changxin Zhang

Publications and source records attributed to Changxin Zhang.

3 recordsLinked to original sources

arXiv2026

Distributed Secure Learning Control for Large-scale Multirobots under Stealthy Actuator Attacks

Distributed learning control for multirobot systems (MRS) offers significant flexibility in presence of uncertainties but lacks provable performance guarantees. A promising direction involves integrating reinforcement learning (RL) into distributed model predictive control (DMPC), leveraging the strengths of RL in nonlinear policy design and the receding-horizon replanning capabilities of DMPC. However, ensuring secure control within such a learning framework under malicious cyber attacks, particularly stealthy ones, remains a critical challenge, because the distributed policies generation depends on information exchange among neighbors, where compromised agents can rapidly influence the behavior of others through the communication network. This article proposes a distributed secure learning control (DSLC) framework for large-scale MRS under malicious, stealthy actuator attacks. Our framework offers two key features: (i) a unified approach that enables secure learning control across various coordination scenarios and (ii) a game-theoretic distributed learning-based predictive control strategy that learns how to balance the attacker and defender through a differential-game based DMPC framework. Specifically, DSLC employs a distributed attacker-actor-critic architecture to learn the optimal defense and attack policies online within each prediction interval. Unlike numerical optimization-based controllers that calculate open-loop control sequences, our method simultaneously generates adversarial attack policies and corresponding defense policies in analytical closed-loop form. The defense policies could be directly generalized to MRS with varying scales and diverse actuator attack probabilities. The effectiveness and scalability of DSLC are validated through comprehensive simulations and real-world experiments in multiple wheeled robots via various control tasks.

cs.RO↗

arXiv2025

Diffusion Policies with Value-Conditional Optimization for Offline Reinforcement Learning

In offline reinforcement learning, value overestimation caused by out-of-distribution (OOD) actions significantly limits policy performance. Recently, diffusion models have been leveraged for their strong distribution-matching capabilities, enforcing conservatism through behavior policy constraints. However, existing methods often apply indiscriminate regularization to redundant actions in low-quality datasets, resulting in excessive conservatism and an imbalance between the expressiveness and efficiency of diffusion modeling. To address these issues, we propose DIffusion policies with Value-conditional Optimization (DIVO), a novel approach that leverages diffusion models to generate high-quality, broadly covered in-distribution state-action samples while facilitating efficient policy improvement. Specifically, DIVO introduces a binary-weighted mechanism that utilizes the advantage values of actions in the offline dataset to guide diffusion model training. This enables a more precise alignment with the dataset's distribution while selectively expanding the boundaries of high-advantage actions. During policy improvement, DIVO dynamically filters high-return-potential actions from the diffusion model, effectively guiding the learned policy toward better performance. This approach achieves a critical balance between conservatism and explorability in offline RL. We evaluate DIVO on the D4RL benchmark and compare it against state-of-the-art baselines. Empirical results demonstrate that DIVO achieves superior performance, delivering significant improvements in average returns across locomotion tasks and outperforming existing methods in the challenging AntMaze domain, where sparse rewards pose a major difficulty.

cs.LG↗

arXiv2023

Centralized Management of a Wifi Mesh for Autonomous Farms

Emerging autonomous farming techniques rely on smart devices such as multi-spectral cameras, collecting fine-grained data, and robots performing tasks such as de-weeding, berry-picking, etc. These techniques require a high throughput network, supporting 10s of Mbps per device at the scale of tens to hundreds of devices in a large farm. We conduct a survey across 12 agronomists to understand these networking requirements of farm workloads and perform extensive measurements of WiFi 6 performance in a farm to identify the challenges in meeting them. Our measurements reveal how network capacity is fundamentally limited in such a setting, with severe degradation in network performance due to crop canopy, and spotlight farm networks as an emerging new problem domain that can benefit from smarter network resource management decisions. To that end, we design Cornet, a network for supporting on-farm applications that comprises: (i) a multi-hop mesh of WiFi routers that uses a strategic combination of 2.4GHz and 5GHz bands as informed by our measurements, and (ii) a centralized traffic engineering (TE) system that uses a novel abstraction of resource units to reason about wireless network capacity and make TE decisions (schedule flows, assign flow rates, and select routes and channels). Our evaluation, using testbeds in a farm and trace-driven simulations, shows how Cornet achieves 1.4 $\times$ higher network utilization and better meets application demands, compared to standard wireless mesh strategies.

cs.NI↗