arXiv subjects

Zixin Jiang

Publications and source records attributed to Zixin Jiang.

11 recordsLinked to original sources

arXiv2026

AE-UAV: An Air-to-Air Event-Based UAV Tracking Benchmark and a Real-Time Frequency-Domain Tracker

Air-to-air (A2A) unmanned aerial vehicle (UAV) tracking is fundamental to airborne remote sensing of low-altitude aerial targets. However, the deployment of continuous, real-time tracking systems on UAVs presents significant challenges. In A2A scenarios, traditional frame-based cameras suffer from severe performance degradation under low illumination, overexposure, and high-speed motion owing to their limited dynamic range and fixed temporal sampling. Although event cameras offer a promising alternative with microsecond temporal resolution and a high dynamic range, current research is bottlenecked by two primary issues: 1) the absence of dedicated A2A event-based datasets, and 2) the heavy reliance of existing trackers on GPU acceleration and extensive training data, rendering them impractical for resource-constrained UAVs. To bridge these gaps, we introduce AE-UAV, an air-to-air event-based UAV tracking benchmark. To the best of our knowledge, this is the first airborne-captured event camera dataset for A2A tracking, comprising 178 flight sequences with continuous-time cubic B-spline annotations. Furthermore, we propose the Fast-Slow Frequency-domain Tracking (FSFT) method. This lightweight, training-free framework seamlessly integrates frequency-domain template matching with search region prediction and detection-based drift correction. Extensive experiments demonstrate that FSFT operates at an ultra-fast 420 frames per second (FPS) on CPU-only hardware. It retains 93.97% of the accuracy of state-of-the-art GPU-dependent methods while delivering a 5.32-fold effective speedup and exhibiting superior temporal resolution generalization, thereby providing a highly efficient and robust solution for airborne remote sensing of aerial targets. The dataset and source code are available at https://github.com/MSP-xEN/AE-UAV.

cs.CV↗

arXiv2026

Buildrix: An Open Platform for Sharing and Benchmarking Agentic AI Skills in Building Engineering

Agentic AI offers significant potential to automate complex building-engineering workflows. However, most existing applications remain isolated proof-of-concept demonstrations and lack reusable domain capabilities, human-verified evaluation cases, and standardized benchmarking infrastructure. This study presents Buildrix, an open, community-driven platform for developing, sharing, executing, and evaluating agentic AI skills for building engineering. Buildrix integrates three components: a Python command-line package for developing, validating, publishing, installing, and managing skills and test cases; a web-based Hub for organizing open challenges, reusable skills, test cases, reviews, and benchmark results; and a local agent harness that supports skill discovery, external toolchain provisioning, progressive context loading, and multi-step workflow execution. Buildrix skills are organized as standardized, self-contained packages containing task instructions, executable scripts, dependencies, and supporting resources. Quantitative test cases can be verified by domain experts and promoted to golden test cases for reproducible benchmark evaluation. Buildrix provides an open foundation for reusable capability development, transparent evaluation, and community-driven advancement of agentic AI in building engineering.

eess.SY↗

arXiv2026

Agents' Last Exam

Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deployment across many professional domains. We argue that this gap is largely an evaluation problem: widely used benchmarks lack sustained performance measurement on real and economically valuable workflows. This paper introduces Agents' Last Exam (ALE), a benchmark designed to evaluate AI agents on long horizon, economically valuable, real world tasks with verifiable outcomes. Developed in collaboration with 250+ industry experts, ALE covers non-physical industries defined with reference to O*NET / SOC 2018 (the U.S. federal occupational taxonomy). It is organized around a task taxonomy with 55 sub fields grouped into 13 industry clusters covering 1K+ tasks. Current results show that the hardest tier remains far from saturated: across mainstream harness and backbone configurations, the average full pass rate is below 1%. ALE is designed as a living benchmark: its task pool grows continuously as new workflows and industries are onboarded. More broadly, ALE is intended not merely as another leaderboard, but as an instrument for closing the gap between benchmark success and GDP relevant impact.

cs.AI↗

arXiv2026

SFCo-Nav: Efficient Zero-Shot Visual Language Navigation via Collaboration of Slow LLM and Fast Attributed Graph Alignment

Recent advances in large vision-language models (VLMs) and large language models (LLMs) have enabled zero-shot approaches to visual language navigation (VLN), where an agent follows natural language instructions using only ego perception and reasoning. However, existing zero-shot methods typically construct a naive observation graph and perform per-step VLM-LLM inference on it, resulting in high latency and computation costs that limit real-time deployment. To address this, we present SFCo-Nav, an efficient zero-shot VLN framework inspired by the principle of slow-fast cognitive collaboration. SFCo-Nav integrates three key modules: 1) a slow LLM-based planner that produces a strategic chain of subgoals, each linked to an imagined object graph; 2) a fast reactive navigator for real-time object graph construction and subgoal execution; and 3) a lightweight asynchronous slow-fast bridge aligns advanced structured, attributed imagined and perceived graphs to estimate navigation confidence, triggering the slow LLM planner only when necessary. To the best of our knowledge, SFCo-Nav is the first slow-fast collaboration zero-shot VLN system supporting asynchronous LLM triggering according to the internal confidence. Evaluated on the public R2R and REVERIE benchmarks, SFCo-Nav matches or exceeds prior state-of-the-art zero-shot VLN success rates while cutting total token consumption per trajectory by over 50% and running more than 3.5 times faster. Finally, we demonstrate SFCo-Nav on a legged robot in a hotel suite, showcasing its efficiency and practicality in indoor environments.

cs.RO↗

arXiv2026

OptAgent: an Agentic AI framework for Intelligent Building Operations

The urgent need for building decarbonization calls for a paradigm shift in future autonomous building energy operation, from human-intensive engineering workflows toward intelligent agents that interact with physics-grounded digital environments. This study proposes an end-to-end agentic AI-enabled Physics-Informed Machine Learning (PIML) environment for scalable building energy modeling, simulation, control, and automation. The framework consists of (1) a modular and physics-consistent PIML digital environment spanning building thermal dynamics, Heating, Ventilation, and Air Conditioning (HVAC), and distributed energy resources (DER) for grid-interactive energy management; and (2) an agentic AI layer with 11 specialist agents and 72 Model Context Protocol (MCP) tools that enable end-to-end execution of multi-step energy analytics. A representative case study demonstrates multi-domain, multi-agent coordination for assessing how system and control upgrades affect energy use, operating cost, thermal comfort, and flexibility. In addition, a large-scale benchmark (about 4000 runs) systematically evaluates workflow performance in terms of accuracy, token consumption, execution time, and inference cost. The results quantify the impacts of intelligence mode design, model size, task complexity, and orchestrator-specialist coordination, and provide key lessons for building future agentic AI systems in real-world building energy applications. This work establishes a scalable, physics-grounded foundation for deploying agentic AI in decarbonized and grid-interactive building operations.

eess.SY↗

arXiv2026

BESTOpt: A Modular, Physics-Informed Machine Learning based Building Modeling, Control and Optimization Framework

Modern buildings are increasingly interconnected with occupancy, heating, ventilation, and air-conditioning (HVAC) systems, distributed energy resources (DERs), and power grids. Modeling, control, and optimization of such multi-domain systems play a critical role in achieving building-sector decarbonization. However, most existing tools lack scalability and physical consistency for addressing these complex, multi-scale ecosystem problems. To bridge this gap, this study presents BESTOpt, a modular, physics-informed machine learning (PIML) framework that unifies building applications, including benchmarking, evaluation, diagnostics, control, optimization, and performance simulation. The framework adopts a cluster-domain-system/building-component hierarchy and a standardized state-action-disturbance-observation data typology. By embedding physics priors into data-driven modules, BESTOpt improves model accuracy and physical consistency under unseen conditions. Case studies on single-building and cluster scenarios demonstrate its capability for multi-level centralized and decentralized control. Looking ahead, BESTOpt lays the foundation for an open, extensible platform that accelerates interdisciplinary research toward smart, resilient, and decarbonized building ecosystems.

eess.SY↗

arXiv2025

Physics-informed Modularized Neural Network for Advanced Building Control by Deep Reinforcement Learning

Physics-informed machine learning (PIML) provides a promising solution for building energy modeling and can serve as a virtual environment to enable reinforcement learning (RL) agents to interact and learn. However, challenges remain in efficiently integrating physics priors, evaluating the effectiveness of physics constraints, balancing model accuracy and physics consistency, and enabling real-world implementation. To address these gaps, this study introduces a Physics-Informed Modularized Neural Network (PI-ModNN), which incorporates physics priors through a physics-informed model structure, loss functions, and hard constraints. A new evaluation metric called "temperature response violation" is developed to quantify the physical consistency of data-driven building dynamic models under varying control inputs and training data sizes. Additionally, a physics prior evaluation framework based on rule importance is proposed to assess the contribution of each individual physics prior, offering guidance on selecting appropriate PIML techniques. Results indicate that incorporating physical priors does not always improve model performance; inappropriate priors may decrease model accuracy and consistency. However, hard constraints are effective in enforcing model consistency. Furthermore, we present a general workflow for developing control-oriented PIML models and integrating them with deep reinforcement learning (DRL). Following this framework, a case study implementing DRL in an office space over three months demonstrates potential energy savings of 31.4%. Finally, we provide a general guideline for integrating data-driven models with advanced building control through a four-step evaluation framework, paving the way for reliable and scalable deployment of advanced building controls.

eess.SY↗

arXiv2025

Physics-informed machine learning for building performance simulation-A review of a nascent field

Building performance simulation (BPS) is critical for understanding building dynamics and behavior, analyzing performance of the built environment, optimizing energy efficiency, improving demand flexibility, and enhancing building resilience. However, conducting BPS is not trivial. Traditional BPS relies on an accurate building energy model, mostly physics-based, which depends heavily on detailed building information, expert knowledge, and case-by-case model calibrations, thereby significantly limiting their scalability. With the development of sensing technology and increased data availability, there is a growing attention and interest in data-driven BPS. However, purely data-driven models often suffer from limited generalization ability and a lack of physical consistency, resulting in poor performance in real-world applications. To address these limitations, recent studies have started to incorporate physics priors into data-driven models, a methodology called physics-informed machine learning (PIML). PIML is an emerging field with the definitions, methodologies, evaluation criteria, application scenarios, and future directions that remain open. To bridge those gaps, this study systematically reviews the state-of-art PIML for BPS, offering a comprehensive definition of PIML, and comparing it to traditional BPS approaches regarding data requirements, modeling effort, performance and computation cost. We also summarize the commonly used methodologies, validation approaches, application domains, available data sources, open-source packages and testbeds. In addition, this study provides a general guideline for selecting appropriate PIML models based on BPS applications. Finally, this study identifies key challenges and outlines future research directions, providing a solid foundation and valuable insights to advance R&D of PIML in BPS.

eess.SY↗

arXiv2024

Modularized Neural Network Incorporating Physical Priors for Smart Building Control, Accuracy or Consistency?

Model predictive control can achieve significant energy savings, offer grid flexibility, and mitigate carbon emissions. However, the challenge of identifying individual control-oriented building dynamic models limits large-scale real-world applications. To address this issue, this study proposed a Modularized Neural Network Incorporating Physical Priors (ModNN), capable of establishing a control-oriented and physical-consistent building dynamic model within minutes without substantial modeling effort. This is also the first study to evaluate the physical consistency of a given data-driven model both qualitatively and quantitively. We compared the physical consistency of a classical Long Short-Term Memory (LSTM) model and our ModNN. The ModNN strictly satisfies physical constraints, whereas the LSTM model learned contradictory system dynamics. Additionally, we compared their control performance on an EnergyPlus virtual testbed. While the LSTM model demonstrated slightly better prediction accuracy in dynamic modeling, it failed in control optimization, resulting in an 89{\deg}C-h temperature violation, whereas the ModNN showed only a 0.57{\deg}C-h violation and achieved up to a 78% peak load reduction. Our findings highlight the importance of incorporating physics priors into data-driven models and provide a promising solution for future smart building control optimization. Furthermore, the proposed evaluation framework defines two physical consistency indicators, providing guidelines for selecting and testing control-oriented, data-driven building dynamic models.

eess.SY↗

arXiv2024

The Smart Buildings Control Suite: A Diverse Open Source Benchmark to Evaluate and Scale HVAC Control Policies for Sustainability

Commercial buildings account for 17% of U.S. carbon emissions, with roughly half of that from Heating, Ventilation, and Air Conditioning (HVAC). HVAC devices form a complex thermodynamic system, and while Model Predictive Control and Reinforcement Learning have been used to optimize control policies, scaling to thousands of buildings remains a significant unsolved challenge. Most current algorithms are over-optimized for specific buildings and rely on proprietary data or hard-to-configure simulations. We present the Smart Buildings Control Suite, the first open source interactive HVAC control benchmark with a focus on solutions that scale. It consists of 3 components: real-world telemetric data extracted from 11 buildings over 6 years, a lightweight data-driven simulator for each building, and a modular Physically Informed Neural Network (PINN) building model as a simulator alternative. The buildings span a variety of climates, management systems, and sizes, and both the simulator and PINN easily scale to new buildings, ensuring solutions using this benchmark are robust to these factors and only reliant on fully scalable building models. This represents a major step towards scaling HVAC optimization from the lab to buildings everywhere. To facilitate use, our benchmark is compatible with the Gym standard, and our data is part of TensorFlow Datasets.

cs.AI↗

arXiv2016

Smoothness of Minkowski sum and generic rotations

Can the Minkowski sum of two compact convex bodies be made smoother by rotating one of them? We construct two infinitely differentiable strictly convex plane bodies such that after any generic rotation (in the Baire category sense) of one of the summands the Minkowski sum is not five times differentiable. On the other hand, if for one of the bodies the zero set of the Gaussian curvature has countable spherical image, we show that any generic rotation makes their Minkowski sum as smooth as the summands. We also improve and clarify some previous results on smoothness of the Minkowski sum.

math.DG↗