HSHyunki Seong

Hyunki Seong

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Hyunki Seong has a total of 36 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Enhancing Robustness in End-to-End Autonomous Driving through a Feedback-based Network and Maneuver-Level Data Augmentation

    by Hyunki Seong, Seongwoo Moon, Jehun Kang, Hojin Ahn, David Hyunchul Shim, Jooyoung Kim, Jaeyun Cho, Junwoo Jung

  • Open Access

    Post-Training and Test-Time Scaling of Generative Agent Behavior Models for Interactive Autonomous Driving

    by Hyunki Seong, Jeong-Kyun Lee, Heesoo Myeong, Yongho Shin, Hyun-Mook Cho, Duck Hoon Kim, Pranav Desai, Monu Surana

    Learning interactive motion behaviors among multiple agents is a core challenge in autonomous driving. While imitation learning models generate realistic trajectories, they often inherit biases from datasets dominated by safe demonstrations, limiting robustness in safety-critical cases. Moreover, most studies rely on open-loop evaluation, overlooking compounding errors in closed-loop execution. We address these limitations with two complementary strategies. First, we propose Group Relative Behavior Optimization (GRBO), a reinforcement learning post-training method that fine-tunes pretrained behavior models via group relative advantage maximization with human regularization. Using only 10% of the training dataset, GRBO improves safety performance by over 40% while preserving behavioral realism. Second, we introduce Warm-K, a warm-started Top-K sampling strategy that balances consistency and diversity in motion selection. Our Warm-K method-based test-time scaling enhances behavioral consistency and reactivity at test time without retraining, mitigating covariate shift and reducing performance discrepancies. Demo videos are available in the supplementary material.

  • Open Access

    VLA-R: Vision-Language Action Retrieval toward Open-World End-to-End Autonomous Driving

    by Hyunki Seong, Seongwoo Moon, Hojin Ahn, Jehun Kang, David Hyunchul Shim

    Exploring open-world situations in an end-to-end manner is a promising yet challenging task due to the need for strong generalization capabilities. In particular, end-to-end autonomous driving in unstructured outdoor environments often encounters conditions that were unfamiliar during training. In this work, we present Vision-Language Action Retrieval (VLA-R), an open-world end-to-end autonomous driving (OW-E2EAD) framework that integrates open-world perception with a novel vision-action retrieval paradigm. We leverage a frozen vision-language model for open-world detection and segmentation to obtain multi-scale, prompt-guided, and interpretable perception features without domain-specific tuning. A Q-Former bottleneck aggregates fine-grained visual representations with language-aligned visual features, bridging perception and action domains. To learn transferable driving behaviors, we introduce a vision-action contrastive learning scheme that aligns vision-language and action embeddings for effective open-world reasoning and action retrieval. Our experiments on a real-world robotic platform demonstrate strong generalization and exploratory performance in unstructured, unseen environments, even with limited data. Demo videos are provided in the supplementary material.

  • Coordinated Multi-Robot Platooning via Prioritized Velocity Planning

    by Hyunki Seong, Seongwoo Moon, D. Hyunchul Shim, Wooyeol Kim, Sunwoo Baek

  • Autonomous System for Head-to-Head Race: Design, Implementation, and Analysis; Team KAIST at the Indy Autonomous Challenge

    by Calvin Chanyoung Chung, Andrea Finazzi, Hyunki Seong, Daegyu Lee, Seungwook Lee, Bosung Kim, Gyuri Gang, David Hyunchul Shim

    Autonomous racing has emerged as a promising field of research, attracting growing interest. In racing, vehicles operate at the limits of perception, planning, and control, raising unique research and engineering challenges. This paper presents an overview of the autonomous racing system developed by team KAIST for the Indy Autonomous Challenge (IAC). The KAIST racing stack features multi-modal perception modules, a high-speed overtaking planner, a resilient control system, and a system monitoring framework. We provide detailed insights into each component, including the algorithms, implementation, and evaluation results. Furthermore, the paper presents design principles shaped by three years of high-speed autonomous racing experience, offering insights applicable to other safety-critical, high-risk robotics domains. Our autonomous system was integrated into a full-scale race car (Dallara AV-21) and extensively field-tested. As a result, team KAIST was one of only three teams to qualify for and compete in the official IAC races without incident. Our system successfully fulfilled all race objectives, including overtaking at speeds of up to 220 km / h in the IAC@CES2022, the world’s first autonomous head-to-head race.


  • TempFuser: Learning Agile, Tactical, and Acrobatic Flight Maneuvers Using a Long Short-Term Temporal Fusion Transformer

    by Hyunki Seong, David Hyunchul Shim

    Dogfighting is a challenging scenario in aerial applications that requires a comprehensive understanding of both strategic maneuvers and the aerodynamics of agile aircraft. The aerial agent needs to not only understand tactically evolving maneuvers of fighter jets from a long-term perspective but also react to rapidly changing aerodynamics of aircraft from a short-term viewpoint. In this paper, we introduce TempFuser, a novel long short-term temporal fusion transformer architecture that can learn agile, tactical, and acrobatic flight maneuvers in complex dogfight problems. Our approach integrates two distinct temporal transition embeddings into a transformer-based network to comprehensively capture both the long-term tactics and short-term agility of aerial agents. By incorporating these perspectives, our policy network generates end-to-end flight commands that secure dominant positions over the long term and effectively outmaneuver agile opponents. After training in a high-fidelity flight simulator, our model successfully learns to execute strategic maneuvers, outperforming baseline policy models against various types of opponent aircraft. Notably, our model exhibits human- like acrobatic maneuvers even when facing adversaries with superior specifications, all without relying on prior knowledge. Moreover, it demonstrates robust pursuit performance in challenging supersonic and low-altitude situations. Demo videos are available at https://sites.google.com/view/tempfuser .


  • Design, Field Evaluation, and Traffic Analysis of a Competitive Autonomous Driving Model in a Congested Environment

    by Daegyu Lee, Hyunki Seong, Gyuree Kang, Seungil Han, D. Hyunchul Shim, Yoonjin Yoon

    Recently, numerous studies have investigated cooperative traffic systems using the communication among vehicle-to-everything (V2X). Unfortunately, when multiple autonomous vehicles are deployed while exposed to communication failure, there might be a conflict of ideal conditions between various autonomous vehicles leading to adversarial situation on the roads. In South Korea, virtual and real-world urban autonomous multi-vehicle races were held in March and November of 2021, respectively. During the competition, multiple vehicles were involved simultaneously, which required maneuvers such as overtaking low-speed vehicles, negotiating intersections, and obeying traffic laws. In this study, we introduce a fully autonomous driving software stack to deploy a competitive driving model, which enabled us to win the urban autonomous multi-vehicle races. We evaluate module-based systems such as navigation, perception, and planning in real and virtual environments. Additionally, an analysis of traffic is performed after collecting multiple vehicle position data over communication to gain additional insight into a multi-agent autonomous driving scenario. Finally, we propose a method for analyzing traffic in order to compare the spatial distribution of multiple autonomous vehicles. We study the similarity distribution between each team’s driving log data to determine the impact of competitive autonomous driving on the traffic environment. Our fully autonomous software architecture, proven successful in winning urban autonomous multi-vehicle races in South Korea, is ready for deployment on urban robot taxis. Our traffic analysis addresses multi-agent scenarios and resolves competitive conflicts among robot taxi companies, crucial for smart city integration and optimizing autonomous vehicle performance in complex urban settings.


  • Open Access

    Model Parameter Identification via a Hyperparameter Optimization Scheme for Autonomous Racing Systems

    by Hyunki Seong, Chanyoung Chung, David Hyunchul Shim

    In this letter, we propose a model parameter identification method via a hyperparameter optimization scheme (MI-HPO). Our method adopts an efficient explore-exploit strategy to identify the parameters of dynamic models in a data-driven optimization manner. We utilize our method for model parameter identification of the AV-21, a full-scaled autonomous race vehicle. We then incorporate the optimized parameters for the design of model-based planning and control systems of our platform. In experiments, MI-HPO exhibits more than 13 times faster convergence than traditional parameter identification methods. Furthermore, the parametric models learned via MI-HPO demonstrate good fitness to the given datasets and show generalization ability in unseen dynamic scenarios. We further conduct extensive field tests to validate our model-based system, demonstrating stable obstacle avoidance and high-speed driving up to 217 km/h at the Indianapolis Motor Speedway and Las Vegas Motor Speedway. The source code for our work and videos of the tests are available at https://github.com/hynkis/MI-HPO.

  • Open Access

    A Versatile Door Opening System with Mobile Manipulator through Adaptive Position-Force Control and Reinforcement Learning

    by Gyuree Kang, Hyunki Seong, Daegyu Lee, D. Hyunchul Shim

    The ability of robots to navigate through doors is crucial for their effective operation in indoor environments. Consequently, extensive research has been conducted to develop robots capable of opening specific doors. However, the diverse combinations of door handles and opening directions necessitate a more versatile door opening system for robots to successfully operate in real-world environments. In this paper, we propose a mobile manipulator system that can autonomously open various doors without prior knowledge. By using convolutional neural networks, point cloud extraction techniques, and external force measurements during exploratory motion, we obtained information regarding handle types, poses, and door characteristics. Through two different approaches, adaptive position-force control and deep reinforcement learning, we successfully opened doors without precise trajectory or excessive external force. The adaptive position-force control method involves moving the end-effector in the direction of the door opening while responding compliantly to external forces, ensuring safety and manipulator workspace. Meanwhile, the deep reinforcement learning policy minimizes applied forces and eliminates unnecessary movements, enabling stable operation across doors with different poses and widths. The RL-based approach outperforms the adaptive position-force control method in terms of compensating for external forces, ensuring smooth motion, and achieving efficient speed. It reduces the maximum force required by 3.27 times and improves motion smoothness by 1.82 times. However, the non-learning-based adaptive position-force control method demonstrates more versatility in opening a wider range of doors, encompassing revolute doors with four distinct opening directions and varying widths.