arXiv subjects

Yutian Zhang

Publications and source records attributed to Yutian Zhang.

7 recordsLinked to original sources

arXiv2026

FWBC-VLA: Force-Aware Whole-Body Compensation for Contact-Rich Loco-Manipulation

Contact-rich loco-manipulation requires a bridge between semantic action generation and physical interaction control. Existing Vision-language-action (VLA) models generate task-level actions from visual and linguistic observations, but cannot interpret the physical interactions induced by those actions. While the whole-body control (WBC) policy can stabilize the robot, it cannot distinguish task-relevant interaction forces from forces induced by external disturbances during manipulation. Although force/torque sensors provide direct measurements of physical interactions, retrofitting them entails additional hardware costs and substantial integration effort, particularly for platforms not designed with sensor integration in mind. To address this problem, we propose FWBC-VLA, a force-aware framework that bridges task-level VLA action generation and low-level whole-body compensation control for wheeled-legged robots. First, we introduce HSR-Force, a sensorless residual-torque estimator for inferring contact strength and its temporal variation. These contact estimates are then encoded as tokens and injected into the VLA action expert during action decoding, enabling the policy to perceive contact onset, sustained loading, and release. For loco-manipulation tasks, all parameters of the pretrained VLA backbone are fine-tuned on our WL\&Arm Dataset, which comprises more than 5,000 episodes. Moreover, the robot's proprioceptive state, the Jacobian-derived body-frame force estimate, and the estimated contact state are jointly fed into a compensation generator to produce corrective actions. The manipulation-centric actions are subsequently combined with the corrective actions and passed to the WBC policy for execution. Real-world experiments on whiteboard wiping and door opening with a door closer demonstrate the effectiveness of our FWBC-VLA in contact-rich loco-manipulation.

cs.RO↗

arXiv2026

ForeTime-VLA: Causal Future-Token Distillation from a World Action Model for Conveyor-Belt Manipulation

Manipulating moving objects requires a policy to anticipate contact events, yet vision-language-action (VLA) policies are commonly fine-tuned from the current observation alone. World action models (WAMs) learn predictive dynamics, but running a video-scale teacher or explicitly imagining future frames at deployment is costly. We introduce ForeTime-VLA, a dense pi0.5 policy that distills a future-aware, action-equivalent representation from a frozen Fast-WAM-derived teacher while remaining causal at inference. Offline, current and future video latents are compressed into a whitened 64-D target. Online, an eight-frame history encoder predicts this target together with manipulation phase and normalized time-to-transition. Four future tokens and one phase token condition the VLM prefix, while the predicted future and transition horizon condition the action expert. Training retains the original flow-matching action target and adds cosine, relational geometry, phase, time-to-transition, and action-equivalence objectives. On a deduplicated conveyor-belt dataset, we compare 40k-step checkpoints on 768 matched windows per split. Test MAE decreases from 0.134119 to 0.130593 (2.63%; paired-bootstrap 95% CI: 0.82-4.48% improvement), and test L2 decreases by 3.02%, at a 2.46-2.93% latency cost. In quantitative real-robot evaluation, ForeTime-VLA achieves 81.1% stationary and 58.9% slow-moving grasp success, exceeding the next-best reference by 12.2 and 22.2 percentage points, respectively. Across three belt speeds, it completes 44/90 grasps versus 23/90 for pi0.5, including 11/30 versus 2/30 at fast speed. The agreement between offline orientation gains and reduced real-robot contact-pose failures supports causal future-token distillation as an effective way to improve dynamic manipulation without deploying the world-model teacher.

cs.AI↗

arXiv2026

DECOWAM: Decoupled Whole-Body World-Action Model for Legged Mobile Manipulation

Mobile manipulation requires a robot to predict how locomotion and arm motion jointly alter future observations and control. Existing world-action models, developed largely for fixed-base platforms, do not explicitly distinguish camera ego-motion from base and arm actions. Here we introduce DECOWAM, a whole-body world-action model that separates these factors through dedicated conditional interfaces. DECOWAM freezes an adapted FastWAM backbone and trains residual adapters, an action-equivalent future bottleneck distilled from privileged observations, adversarially separated base and arm latents, and base-velocity conditioning for video prediction. We further introduce ARMDOG, a real-robot dataset that synchronizes video, whole-body state and action, and language. On a fixed replay protocol, DECOWAM improved both future-video and action prediction over FastWAM, reducing action MSE by 21.7% with 25.95M trainable adaptation parameters. Across 79 closed-loop trials per method, it achieved the highest observed whole-body coordination and base-displacement robustness among the compared systems, while task completion remained comparable to the strongest baseline. These results show that embodiment-aware factorization can support parameter-efficient joint visual prediction and whole-body control under moving viewpoints.

cs.AI↗

arXiv2026

Boosting AI Reliability with an FSM-Driven Streaming Inference Pipeline: An Industrial Case

The widespread adoption of AI in industry is often hampered by its limited robustness when faced with scenarios absent from training data, leading to prediction bias and vulnerabilities. To address this, we propose a novel streaming inference pipeline that enhances data-driven models by explicitly incorporating prior knowledge. This paper presents the work on an industrial AI application that automatically counts excavator workloads from surveillance videos. Our approach integrates an object detection model with a Finite State Machine (FSM), which encodes knowledge of operational scenarios to guide and correct the AI's predictions on streaming data. In experiments on a real-world dataset of over 7,000 images from 12 site videos, encompassing more than 300 excavator workloads, our method demonstrates superior performance and greater robustness compared to the original solution based on manual heuristic rules. We will release the code at https://github.com/thulab/video-streamling-inference-pipeline.

cs.CV↗

arXiv2024

MARLens: Understanding Multi-agent Reinforcement Learning for Traffic Signal Control via Visual Analytics

The issue of traffic congestion poses a significant obstacle to the development of global cities. One promising solution to tackle this problem is intelligent traffic signal control (TSC). Recently, TSC strategies leveraging reinforcement learning (RL) have garnered attention among researchers. However, the evaluation of these models has primarily relied on fixed metrics like reward and queue length. This limited evaluation approach provides only a narrow view of the model's decision-making process, impeding its practical implementation. Moreover, effective TSC necessitates coordinated actions across multiple intersections. Existing visual analysis solutions fall short when applied in multi-agent settings. In this study, we delve into the challenge of interpretability in multi-agent reinforcement learning (MARL), particularly within the context of TSC. We propose MARLens a visual analytics system tailored to understand MARL-based TSC. Our system serves as a versatile platform for both RL and TSC researchers. It empowers them to explore the model's features from various perspectives, revealing its decision-making processes and shedding light on interactions among different agents. To facilitate quick identification of critical states, we have devised multiple visualization views, complemented by a traffic simulation module that allows users to replay specific training scenarios. To validate the utility of our proposed system, we present three comprehensive case studies, incorporate insights from domain experts through interviews, and conduct a user study. These collective efforts underscore the feasibility and effectiveness of MARLens in enhancing our understanding of MARL-based TSC systems and pave the way for more informed and efficient traffic management strategies.

cs.HC↗

arXiv2024

CSLens: Towards Better Deploying Charging Stations via Visual Analytics -- A Coupled Networks Perspective

In recent years, the global adoption of electric vehicles (EVs) has surged, prompting a corresponding rise in the installation of charging stations. This proliferation has underscored the importance of expediting the deployment of charging infrastructure. Both academia and industry have thus devoted to addressing the charging station location problem (CSLP) to streamline this process. However, prevailing algorithms addressing CSLP are hampered by restrictive assumptions and computational overhead, leading to a dearth of comprehensive evaluations in the spatiotemporal dimensions. Consequently, their practical viability is restricted. Moreover, the placement of charging stations exerts a significant impact on both the road network and the power grid, which necessitates the evaluation of the potential post-deployment impacts on these interconnected networks holistically. In this study, we propose CSLens, a visual analytics system designed to inform charging station deployment decisions through the lens of coupled transportation and power networks. CSLens offers multiple visualizations and interactive features, empowering users to delve into the existing charging station layout, explore alternative deployment solutions, and assess the ensuring impact. To validate the efficacy of CSLens, we conducted two case studies and engaged in interviews with domain experts. Through these efforts, we substantiated the usability and practical utility of CSLens in enhancing the decision-making process surrounding charging station deployment. Our findings underscore CSLens's potential to serve as a valuable asset in navigating the complexities of charging infrastructure planning.

cs.HC↗

arXiv2022

Quantum and Non-local Effects Offer LiDAR over 40dB Advantage

Non-local effects have the potential to radically move forward quantum enhanced LiDAR to provide an advantage over classical LiDAR not only in laboratory environments but practical implementation. In this work, we demonstrate a 43dB lower signal-to-noise ratio using a quantum enhanced LiDAR based on time-frequency entanglement compared with a classical phase-insensitive LiDAR system. Our system can tolerate more than 3 orders of magnitude higher noise than classical singlephoton counting LiDAR systems before detector saturation. To achieve these advantages, we use non-local cancellation of dispersion to take advantage of the strong temporal correlations in photon pairs in spite of the orders of magnitude larger detector temporal uncertainty. We go on to incorporate this scheme with purpose-built scanning collection optics to image non-reflecting targets in an environment with noise.

quant-ph↗