Xiaocong Li

Qing Guo, Xinhang Li, Junyu Chen et al.

Leveraging large language models (LLMs) in traffic signal control (TSC) improves optimization efficiency and interpretability compared to traditional reinforcement learning (RL) methods. However, existing LLM-based approaches are limited by fixed time signal durations and are prone to hallucination errors, while RL methods lack robustness in signal timing decisions and suffer from poor generalization. To address these challenges, this paper proposes HeraldLight, a dual LLMs architecture enhanced by Herald guided prompts. The Herald Module extracts contextual information and forecasts queue lengths for each traffic phase based on real-time conditions. The first LLM, LLM-Agent, uses these forecasts to make fine grained traffic signal control, while the second LLM, LLM-Critic, refines LLM-Agent's outputs, correcting errors and hallucinations. These refined outputs are used for score-based fine-tuning to improve accuracy and robustness. Simulation experiments using CityFlow on real world datasets covering 224 intersections in Jinan (12), Hangzhou (16), and New York (196) demonstrate that HeraldLight outperforms state of the art baselines, achieving a 20.03% reduction in average travel time across all scenarios and a 10.74% reduction in average queue length on the Jinan and Hangzhou scenarios. The source code is available on GitHub: https://github.com/BUPT-ANTlab/HeraldLight.

Hongxuan Wang, Xiaocong Li, Lihao Zheng et al.

Controller tuning and optimization have been among the most fundamental problems in robotics and mechatronic systems. The traditional methodology is usually model-based, but its performance heavily relies on an accurate mathematical system model. In control applications with complex dynamics, obtaining a precise model is often challenging, leading us towards a data-driven approach. While various researchers have explored the optimization of a single controller, it remains a challenge to obtain the optimal controller parameters safely and efficiently when multiple controllers are involved. In this paper, we propose SafeCtrlBO to optimize multiple controllers simultaneously and safely. We simplify the exploration process in safe Bayesian optimization, reducing computational effort without sacrificing expansion capability. Additionally, we use additive kernels to enhance the efficiency of Gaussian process updates for unknown functions. Hardware experimental results on a permanent magnet synchronous motor (PMSM) demonstrate that compared to existing safe Bayesian optimization algorithms, SafeCtrlBO can obtain optimal parameters more efficiently while ensuring safety.

Ruiteng Zhao, Wenshuo Wang, Yicheng Ma et al.

Force sensing is a crucial modality for Vision-Language-Action (VLA) frameworks, as it enables fine-grained perception and dexterous manipulation in contact-rich tasks. We present Force-Distilled VLA (FD-VLA), a novel framework that integrates force awareness into contact-rich manipulation without relying on physical force sensors. The core of our approach is a Force Distillation Module (FDM), which distills force by mapping a learnable query token, conditioned on visual observations and robot states, into a predicted force token aligned with the latent representation of actual force signals. During inference, this distilled force token is injected into the pretrained VLM, enabling force-aware reasoning while preserving the integrity of its vision-language semantics. This design provides two key benefits: first, it allows practical deployment across a wide range of robots that lack expensive or fragile force-torque sensors, thereby reducing hardware cost and complexity; second, the FDM introduces an additional force-vision-state fusion prior to the VLM, which improves cross-modal alignment and enhances perception-action robustness in contact-rich scenarios. Surprisingly, our physical experiments show that the distilled force token outperforms direct sensor force measurements as well as other baselines, which highlights the effectiveness of this force-distilled VLA approach.

Haidong Huang, Haiyue Zhu. Jiayu Song, Xixin Zhao et al.

Offline-to-online reinforcement learning (O2O-RL) has emerged as a promising paradigm for safe and efficient robotic policy deployment but suffers from two fundamental challenges: limited coverage of multimodal behaviors and distributional shifts during online adaptation. We propose UEPO, a unified generative framework inspired by large language model pretraining and fine-tuning strategies. Our contributions are threefold: (1) a multi-seed dynamics-aware diffusion policy that efficiently captures diverse modalities without training multiple models; (2) a dynamic divergence regularization mechanism that enforces physically meaningful policy diversity; and (3) a diffusion-based data augmentation module that enhances dynamics model generalization. On the D4RL benchmark, UEPO achieves +5.9\% absolute improvement over Uni-O4 on locomotion tasks and +12.4\% on dexterous manipulation, demonstrating strong generalization and scalability.

Wenxin Wang, Jun Ma, Zilong Cheng et al.

Mechatronic systems are commonly used in the industry, where fast and accurate motion performance is always required to guarantee manufacturing precision and efficiency. Nevertheless, the system model and parameters are difficult to be obtained accurately. Moreover, the high-order modes, strong coupling in the multi-axis systems, or unmodeled frictions will bring uncertain dynamics to the system. To overcome the above-mentioned issues and enhance the motion performance, this paper introduces a novel intelligent and totally model-free control method for mechatronic systems with unknown dynamics. In detail, a 2-degree-of-freedom (DOF) architecture is designed, which organically merges a generalized super-twisting algorithm with a unique iterative learning law. The controller solely utilizes the input-output data collected in iterations such that it works without any knowledge of the system parameters. The rigorous proof of convergence ability is given and a case study on flexture-joint dual-drive H-gantry stage is shown to validate the effectiveness of the proposed method.

Haiyue Zhu, Yiting Li, Fengjun Bai et al.

Data-efficient domain adaptation with only a few labelled data is desired for many robotic applications, e.g., in grasping detection, the inference skill learned from a grasping dataset is not universal enough to directly apply on various other daily/industrial applications. This paper presents an approach enabling the easy domain adaptation through a novel grasping detection network with confidence-driven semi-supervised learning, where these two components deeply interact with each other. The proposed grasping detection network specially provides a prediction uncertainty estimation mechanism by leveraging on Feature Pyramid Network (FPN), and the mean-teacher semi-supervised learning utilizes such uncertainty information to emphasizing the consistency loss only for those unlabelled data with high confidence, which we referred it as the confidence-driven mean teacher. This approach largely prevents the student model to learn the incorrect/harmful information from the consistency loss, which speeds up the learning progress and improves the model accuracy. Our results show that the proposed network can achieve high success rate on the Cornell grasping dataset, and for domain adaptation with very limited data, the confidence-driven mean teacher outperforms the original mean teacher and direct training by more than 10% in evaluation loss especially for avoiding the overfitting and model diverging.

Xiaocong Li, Haiyue Zhu, Jun Ma et al.

The performance achieved with traditional model-based control system design approaches typically relies heavily upon accurate modeling of the motion dynamics. However, modeling the true dynamics of present-day increasingly complex systems can be an extremely challenging task; and the usually necessary practical approximations often render the automation system to operate in a non-optimal condition. This problem can be greatly aggravated in the case of a multi-axis magnetically-levitated nanopositioning system where the fully floating behavior and multi-axis coupling make extremely accurate identification of the motion dynamics largely impossible. On the other hand, in many related industrial automation applications, e.g., the scanning process with the maglev system, repetitive motions are involved which could generate a large amount of motion data under non-optimal conditions. These motion data essentially contain rich information; therefore, the possibility exists to develop an intelligent automation system to learn from these motion data and to drive the system to operate towards optimality in a data-driven manner. Along this line then, this paper proposes a data-driven controller optimization approach that learns from the past non-optimal motion data to iteratively improve the motion control performance. Specifically, a novel data-driven multi-objective optimization approach is proposed that is able to automatically estimate the gradient and Hessian purely based on the measured motion data; the multi-objective cost function is suitably designed to take into account both smooth and accurate trajectory tracking. Experiments are then conducted on the maglev nanopositioning system to demonstrate the effectiveness of the proposed method, and the results show rather clearly the practical appeal of our methodology for related complex robotic systems with no accurate model available.