Yuxiao Chen

Nvidia (United States)

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Aaron D. Ames

California Institute of Technology

Marco Pavone

Nvidia (United Kingdom)

Boris Ivanovic

Nvidia (United Kingdom)

Dimitris Metaxas

Rutgers, The State University of New Jersey

Huei Peng

University of Michigan–Ann Arbor

Jessy W. Grizzle

University of Michigan–Ann Arbor

L. Zhao

East University Of Heilongjiang

James Anderson

Columbia University

Ugo Rosolia

California Institute of Technology

Andrew Singletary

California Institute of Technology

Cooperative Institutions

China Railway Construction Corporation (China)

285

Weatherford College

114

Istituto Tecnico Industriale Alessandro Volta

Chinese Academy of Sciences

Adrian College

Directorate of Medicinal and Aromatic Plants Research

Stanford University

Google (United States)

University of Hong Kong

Tsinghua University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Preliminary results on correct-by-construction control software synthesis for adaptive cruise control

Petter Nilsson Omar Hussien Yuxiao Chen Ayça Balkan Matthias Rungger

A plethora of driver convenience and safety automation systems are being introduced into production vehicles, such as electronic stability control, adaptive cruise control, lane keeping, and obstacle avoidance. Assuring the seamless and safe integration of each new automation function with existing control functions is a major challenge for vehicle manufacturers. This challenge is compounded by having different suppliers providing software modules for different control functionalities. In this paper, we report on our preliminary steps to address this problem through a fresh perspective combining formal methods, control theory, and correct-by-construction software synthesis. In particular, we begin the process of synthesizing the control software module for adaptive cruise control from formal specifications given in Linear Temporal Logic. In the longer run, we will endow each interacting software module with an assume-guarantee specification stating under which environment assumptions the module is guaranteed to meet its specifications. These assume-guarantee specifications will then be used to formally prove correctness of the cyber-physical system obtained when the integrated modules interact with the physical dynamics.

Cyber-physical system

10.1109/cdc.2014.7039482

Cite

Citations (42)

Language-Guided Traffic Simulation via Scene-Level Diffusion

arXiv (Cornell University) (2023)

Ziyuan Zhong Davis Rempe Yuxiao Chen Boris Ivanovic Yulong Cao

Realistic and controllable traffic simulation is a core capability that is necessary to accelerate autonomous vehicle (AV) development. However, current approaches for controlling learning-based traffic models require significant domain expertise and are difficult for practitioners to use. To remedy this, we present CTG++, a scene-level conditional diffusion model that can be guided by language instructions. Developing this requires tackling two challenges: the need for a realistic and controllable traffic model backbone, and an effective method to interface with a traffic model using language. To address these challenges, we first propose a scene-level diffusion model equipped with a spatio-temporal transformer backbone, which generates realistic and controllable traffic. We then harness a large language model (LLM) to convert a user's query into a loss function, guiding the diffusion model towards query-compliant generation. Through comprehensive evaluation, we demonstrate the effectiveness of our proposed method in generating realistic, query-compliant traffic simulations.

Interface (matter)

10.48550/arxiv.2306.06344

Cite

Citations (3)

Wolf: Captioning Everything with a World Summarization Framework

arXiv (Cornell University) (2024)

Bo-Yi Li Ligeng Zhu Ran Tian Shuhan Tan Yuxiao Chen

We propose Wolf, a WOrLd summarization Framework for accurate video captioning. Wolf is an automated captioning framework that adopts a mixture-of-experts approach, leveraging complementary strengths of Vision Language Models (VLMs). By utilizing both image and video models, our framework captures different levels of information and summarizes them efficiently. Our approach can be applied to enhance video understanding, auto-labeling, and captioning. To evaluate caption quality, we introduce CapScore, an LLM-based metric to assess the similarity and quality of generated captions compared to the ground truth captions. We further build four human-annotated datasets in three domains: autonomous driving, general scenes, and robotics, to facilitate comprehensive comparisons. We show that Wolf achieves superior captioning performance compared to state-of-the-art approaches from the research community (VILA1.5, CogAgent) and commercial solutions (Gemini-Pro-1.5, GPT-4V). For instance, in comparison with GPT-4V, Wolf improves CapScore both quality-wise by 55.6% and similarity-wise by 77.4% on challenging driving videos. Finally, we establish a benchmark for video captioning and introduce a leaderboard, aiming to accelerate advancements in video understanding, captioning, and data alignment. Leaderboard: https://wolfv0.github.io/leaderboard.html.

Closed captioning

10.48550/arxiv.2407.18908

Cite

Citations (0)

Modelling of uncertain reactive human driving behavior: a classification approach

Yuxiao Chen Nauman Sohani Huei Peng

This paper discusses a novel approach to model human driver behavior. A classification-based method is proposed to construct a reactive bound on possible human driving actions given the scenario description (such as the vehicle states and the behavior of surrounding vehicles). This approach captures the reactiveness and uncertainty of human drivers. Real human driving data is used as the positive training data, while dangerous actions sampled via a Hamilton Jacobi reachability computation constitute the negative training data. A classifier that separates the two groups is then trained via a customized L1. Support Vector Machine (SVM), and an analytical bound function is derived from the classifier which maps the state and surrounding vehicles' actions to the bound on possible actions of the human driver. The credibility of the proposed approach is analyzed under the random convex optimization framework. Potential applications of this work include the computation of safe sets, synthesis of safety guaranteed controllers for systems interacting with humans such as autonomous vehicles, and evaluation of such systems.

10.1109/cdc.2018.8619568

Cite

Citations (10)

Investigation of the Imaging Polarization Effect Based on a Pixellated CdZnTe Detector

Chinese Physics Letters (2010)

Miao Li Shali Xiao Liuqiang Zhang Yulin Cao Yuxiao Chen

A pixel array CdZnTe imaging system, employing a 40 × 40 × 5 mm3 pixellated CdZnTe detector, is established. The imaging polarization effect in the CdZnTe pixellated detector for a collimated Cs137 Gamma source is investigated in detail. The experimental results for different irradiated fluxes indicate that excessive irradiated flux indeed causes central pixels to be shut off completely. The imaging performance of the polarized detector is severely degraded. Polarized detector counts are simultaneously reduced to one-third of the non-polarized detector counts. A theoretical model of potential distribution is also proposed by solving the Poisson equation and, in turn, the electric potential distortion for high irradiated flux is discussed by comparison with the experimental results.

Collimated light

10.1088/0256-307x/27/7/070702

Cite

Citations (6)

Optimal Safe Controller Synthesis: A Density Function Approach

2022 American Control Conference (ACC) (2020)

Yuxiao Chen Mohamadreza Ahmadi Aaron D. Ames

This paper considers the synthesis of optimal safe controllers based on density functions. We present an algorithm for robust constrained optimal control synthesis using the duality relationship between the density function and the value function. The density function follows the Liouville equation and is the dual of the value function, which satisfies Bellman's optimality principle. Thanks to density functions, constraints over the distribution of states, such as safety constraints, can be posed straightforwardly in an optimal control problem. The constrained optimal control problem is then solved with a primal-dual algorithm. This formulation is extended to the case with external disturbances, and we show that the robust constrained optimal control can be solved with a modified primal-dual algorithm. We apply this formulation to the problem of finding the optimal safe controller that minimizes the cumulative intervention. An adaptive cruise control (ACC) example is used to demonstrate the efficacy of the proposed, wherein we compare the result of the density function approach with the conventional control barrier function (CBF) method.

Duality (order theory)

10.23919/acc45564.2020.9147721

Cite

Citations (18)

Study of noise distribution of active fiber loop replicator

Optical Technique (2007)

Yuxiao Chen

A numerical model of active fiber loop replicator (AFLR) based on a steady-state model of semiconductor optical amplifier(SOA) is used to analyze the output amplified spontaneous emission(ASE) power distribution when the bias current and input power of active finer loop replicator(AFLR) is changed. An analytical analysis of output noise performance of the system is carried out. An interpolating method to recover original noise distribution using measured data is validated, and the recovery error is measured by an experimental way.

Source

Cite

Citations (2)

Versatile Scene-Consistent Traffic Scenario Generation as Optimization with Diffusion

arXiv (Cornell University) (2024)

Zhiyu Huang Zixu Zhang Ameya Vaidya Yuxiao Chen Chen Lv

Generating realistic and controllable agent behaviors in traffic simulation is crucial for the development of autonomous vehicles. This problem is often formulated as imitation learning (IL) from real-world driving data by either directly predicting future trajectories or inferring cost functions with inverse optimal control. In this paper, we draw a conceptual connection between IL and diffusion-based generative modeling and introduce a novel framework Versatile Behavior Diffusion (VBD) to simulate interactive scenarios with multiple traffic participants. Our model not only generates scene-consistent multi-agent interactions but also enables scenario editing through multi-step guidance and refinement. Experimental evaluations show that VBD achieves state-of-the-art performance on the Waymo Sim Agents benchmark. In addition, we illustrate the versatility of our model by adapting it to various applications. VBD is capable of producing scenarios conditioning on priors, integrating with model-based optimization, sampling multi-modal scene-consistent scenarios by fusing marginal predictions, and generating safety-critical scenarios when combined with a game-theoretic solver.

10.48550/arxiv.2404.02524

Cite

Citations (0)

Interpretable Trajectory Prediction for Autonomous Vehicles via Counterfactual Responsibility

2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (2023)

Kai–Chieh Hsu Karen Leung Yuxiao Chen Jaime F. Fisac Marco Pavone

The ability to anticipate surrounding agents' behaviors is critical to enable safe and seamless autonomous vehicles (AVs). While phenomenological methods have successfully predicted future trajectories from scene context, these predictions lack interpretability. On the other hand, ontological approaches assume an underlying structure able to describe the interaction dynamics or agents' internal decision processes. Still, they often suffer from poor scalability or cannot reflect diverse human behaviors. This work proposes an interpretability framework for a phenomenological method through responsibility evaluations. We formulate responsibility as a measure of how much an agent takes into account the welfare of other agents through counterfactual reasoning. Additionally, this framework abstracts the computed responsibility sequences into different responsibility levels and grounds these latent levels into reward functions. The proposed responsibility-based interpretability framework is modular and easily integrated into a wide range of prediction models. To demonstrate the utility of the proposed framework in providing added interpretability, we adapt an existing AV prediction model and perform a simulation study on a real-world nuScenes traffic dataset. Experimental results show that we can perform offline ex-post traffic analysis by incorporating the responsibility signal and rendering interpretable but accurate online trajectory predictions.

Interpretability

10.1109/iros55552.2023.10341712

Cite

Citations (1)

Tree-structured Policy Planning with Learned Behavior Models

arXiv (Cornell University) (2023)

Yuxiao Chen Péter Karkus Boris Ivanovic Xinshuo Weng Marco Pavone

Autonomous vehicles (AVs) need to reason about the multimodal behavior of neighboring agents while planning their own motion. Many existing trajectory planners seek a single trajectory that performs well under \emph{all} plausible futures simultaneously, ignoring bi-directional interactions and thus leading to overly conservative plans. Policy planning, whereby the ego agent plans a policy that reacts to the environment's multimodal behavior, is a promising direction as it can account for the action-reaction interactions between the AV and the environment. However, most existing policy planners do not scale to the complexity of real autonomous vehicle applications: they are either not compatible with modern deep learning prediction models, not interpretable, or not able to generate high quality trajectories. To fill this gap, we propose Tree Policy Planning (TPP), a policy planner that is compatible with state-of-the-art deep learning prediction models, generates multistage motion plans, and accounts for the influence of ego agent on the environment behavior. The key idea of TPP is to reduce the continuous optimization problem into a tractable discrete Markov Decision Process (MDP) through the construction of two tree structures: an ego trajectory tree for ego trajectory options, and a scenario tree for multi-modal ego-conditioned environment predictions. We demonstrate the efficacy of TPP in closed-loop simulations based on real-world nuScenes dataset and results show that TPP scales to realistic AV scenarios and significantly outperforms non-policy baselines.

Planner

Tree (set theory)

10.48550/arxiv.2301.11902

Cite

Citations (0)