Avatar

Hayoung Kim

Ph.D. Student at Automotive Engineering

Hanyang University

Biography

I am a Ph.D. student in automotive engineering at Hanyang University, advised by prof. Kunsoo Huh in Machine Monitoring and Control (MMC) Lab. My research focuses on interaction aware motion planning, trajectory prediction and estimation in autonomous driving.

Interests

  • Motion Planning under Uncertainty
  • Trajectory Prediction
  • Estimation

Education

  • PhD in Automotive Engineering, Expected 2021 spring

    Hanyang University

  • BS in Mechanical Engineering, 2014

    Hanyang University

Publications

Vehicle sideslip angle estimation using deep ensemble-based adaptive Kalman filter

A novel sideslip angle estimation method combining neural network and Kalman filter

Interaction Aware Trajectory Prediction of Surrounding Vehicles with Interaction Network and Deep Ensemble

For the path planning of autonomous vehicles, it is important to predict the future trajectory of the surrounding vehicles. However, …

Multi-Head Attention-based Probabilistic Vehicle Trajectory Prediction

This paper presents online-capable deep learning model for probabilistic vehicle trajectory prediction. We propose a simple …

Anomaly Monitoring Framework in Lane Detection With a Generative Adversarial Network

Overcoming data imbalance problem in anomaly detection through GAN for commercial lane detection sensor

Action Conditioned Response Prediction with Uncertainty for Automated Vehicles

Human driver modeling for response prediction given the planned future actions of automated vehicle, Best student paper award 1st place

Experience

 
 
 
 
 

Research Intern

Hyundai Motor Company

Jun 2020 – Aug 2020 Seoul, Korea

Develop path planning algorithm in various tight environments (left/right/U turn).

Implement fast collision check algorithm comparing polygon and occupancy grid map.

Implement conformal lattice planner explicitly considering vehicle model with steering limits.

Develop uncertainty-aware planning algorithm by approximating covariance with polytopes.

Contact