Changhee Kim
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Changhee Kim has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Kyongsu Yi
Department of Mechanical Engineering, Seoul National University, Seoul, South Korea
5 shared publications - Youngmin Yoon
Future Mobility Technology Center (FMTC), Seoul National University (SNU), Siheung, South Korea
3 shared publications Dabin Seo
Department of Mechanical Engineering, Seoul National University, Seoul, South Korea
1 shared publication- Donghoon Shin
Division of Artificial Intelligence Engineering, Korea Maritime and Ocean University, Busan, Republic of Korea
1 shared publication - Eunhyek Joa
Model Predictive Control Lab, Department of Mechanical Engineering, University of California at Berkeley, Berkeley, CA, USA
1 shared publication - Heeseong Lee
Department of Mechanical Engineering, Seoul National University, Seoul, South Korea
1 shared publication
Latest Publications
- 11 June 2025
- Published by Springer Nature in International Journal of Control, Automation, and Systems
- Vol. 23 (6) , 1613-1629
Vehicle Motion Planning and Control at the Limits of Handling for Autonomous Driving on Racetrack Environments
- 11 June 2025
- Published by Springer Nature in International Journal of Control, Automation, and Systems
- Vol. 23 (6) , 1613-1629
Vehicle Motion Planning and Control at the Limits of Handling for Autonomous Driving on Racetrack Environments
- 5 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems Letters
- Vol. 8, 2187-2192
Piecewise Affine Relaxation of Discrete Value Functions in Learning Model Predictive Control With Application to Autonomous Racing
Learning Model Predictive Control (LMPC) is a data-driven approach to MPC that enhances closed-loop performance by leveraging data from successive task iterations to approximate the solution of optimal control problems. The value function in LMPC is pivotal for performance enhancement, but its discrete nature—where each point corresponds to a data point—renders the LMPC problem computationally intensive due to its mixed-integer nature. This letter introduces a novel method to construct the LMPC value function. The proposed value function is a piecewise affine approximation that interpolates the discrete data points of the original value function, resulting in a nonlinear relaxation of the mixed-integer LMPC problem. By connecting the discrete data points with piecewise affine segments, the essential characteristics of the original value function are preserved. The proposed algorithm’s effectiveness is demonstrated through numerical simulations in autonomous racing.
- 5 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems Letters
- Vol. 8, 2187-2192
Piecewise Affine Relaxation of Discrete Value Functions in Learning Model Predictive Control With Application to Autonomous Racing
Learning Model Predictive Control (LMPC) is a data-driven approach to MPC that enhances closed-loop performance by leveraging data from successive task iterations to approximate the solution of optimal control problems. The value function in LMPC is pivotal for performance enhancement, but its discrete nature—where each point corresponds to a data point—renders the LMPC problem computationally intensive due to its mixed-integer nature. This letter introduces a novel method to construct the LMPC value function. The proposed value function is a piecewise affine approximation that interpolates the discrete data points of the original value function, resulting in a nonlinear relaxation of the mixed-integer LMPC problem. By connecting the discrete data points with piecewise affine segments, the essential characteristics of the original value function are preserved. The proposed algorithm’s effectiveness is demonstrated through numerical simulations in autonomous racing.
- 24 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (10) , 13163-13183
Spatio-Temporal Corridor-Based Motion Planning of Lane Change Maneuver for Autonomous Driving in Multi-Vehicle Traffic
This paper presents a methodology of lane change motion planning based on spatio-temporal corridor for autonomous driving in multi-vehicle traffic environments. The spatio-temporal corridor is constructed based on the prediction results of surrounding vehicles. For lane change decision making, the target lane change space is determined by investigating the feasibility and space sufficiency using the spatio-temporal corridor. The set of lane crossing time which enables the safe lane change is calculated in the feasibility investigation process. The feasibility investigation secures the consistency of the decision making and motion planning. The lane change motion planning algorithm incorporates decoupled lateral and longitudinal dynamics. The lateral motion is planned based on a sampling method which refers to the lane crossing time set. Subsequently, the longitudinal motion is planned using Model Predictive Control (MPC), taking into account the results of the lateral motion planning. Simulation studies have been conducted to investigate the performance of the proposed lane change algorithm. The proposed algorithm has been implemented in our test vehicle and validated via autonomous driving tests. The vehicle test results show that the proposed algorithm plans the lane change motion safely with proper ride quality.
- 24 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (10) , 13163-13183
Spatio-Temporal Corridor-Based Motion Planning of Lane Change Maneuver for Autonomous Driving in Multi-Vehicle Traffic
This paper presents a methodology of lane change motion planning based on spatio-temporal corridor for autonomous driving in multi-vehicle traffic environments. The spatio-temporal corridor is constructed based on the prediction results of surrounding vehicles. For lane change decision making, the target lane change space is determined by investigating the feasibility and space sufficiency using the spatio-temporal corridor. The set of lane crossing time which enables the safe lane change is calculated in the feasibility investigation process. The feasibility investigation secures the consistency of the decision making and motion planning. The lane change motion planning algorithm incorporates decoupled lateral and longitudinal dynamics. The lateral motion is planned based on a sampling method which refers to the lane crossing time set. Subsequently, the longitudinal motion is planned using Model Predictive Control (MPC), taking into account the results of the lateral motion planning. Simulation studies have been conducted to investigate the performance of the proposed lane change algorithm. The proposed algorithm has been implemented in our test vehicle and validated via autonomous driving tests. The vehicle test results show that the proposed algorithm plans the lane change motion safely with proper ride quality.
- Open Access
- 13 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 5772-5788
Trajectory Planning and Control of Autonomous Vehicles for Static Vehicle Avoidance in Dynamic Traffic Environments
This paper presents a trajectory planning and control algorithm of autonomous vehicles for static traffic agent avoidance in multi vehicle urban environments. In urban autonomous driving, the subject vehicle encounters diverse traffic scenes including lane changing, intersection driving, and illegally parked static vehicle avoidance. Among these, dealing with illegally parked static target vehicle is a major challenge to urban autonomous driving due to large velocity difference between ego and target vehicles and interactions with surrounding vehicles. In order to tackle this problem, we introduce a decision making and motion planning framework for static vehicle avoidance considering both the preceding static vehicles and surrounding vehicles. Among the surrounding vehicles, the set of objects with potential collision risk is selected based on the lane boundaries and road geometry. Then, the driving status of the selected target vehicles are classified as normal driving vehicles or parked vehicles based on their longitudinal speed, lateral position and lateral space occupancy. For the preceding parked vehicles, the motion planner generates lateral and longitudinal evasive motion, by taking side lane traffic flow and risk into account. The desired motion is executed by applying optimized control inputs computed by lateral and longitudinal model predictive controllers. The performance validation of the proposed algorithm has been conducted with actual autonomous test vehicles. The test results confirmed that the proposed algorithm can successfully perform evasive maneuvers on urban roads to ensure safety and mitigate collision risk with the surrounding traffic agents.
- 13 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 5772-5788
Open AccessTrajectory Planning and Control of Autonomous Vehicles for Static Vehicle Avoidance in Dynamic Traffic Environments
This paper presents a trajectory planning and control algorithm of autonomous vehicles for static traffic agent avoidance in multi vehicle urban environments. In urban autonomous driving, the subject vehicle encounters diverse traffic scenes including lane changing, intersection driving, and illegally parked static vehicle avoidance. Among these, dealing with illegally parked static target vehicle is a major challenge to urban autonomous driving due to large velocity difference between ego and target vehicles and interactions with surrounding vehicles. In order to tackle this problem, we introduce a decision making and motion planning framework for static vehicle avoidance considering both the preceding static vehicles and surrounding vehicles. Among the surrounding vehicles, the set of objects with potential collision risk is selected based on the lane boundaries and road geometry. Then, the driving status of the selected target vehicles are classified as normal driving vehicles or parked vehicles based on their longitudinal speed, lateral position and lateral space occupancy. For the preceding parked vehicles, the motion planner generates lateral and longitudinal evasive motion, by taking side lane traffic flow and risk into account. The desired motion is executed by applying optimized control inputs computed by lateral and longitudinal model predictive controllers. The performance validation of the proposed algorithm has been conducted with actual autonomous test vehicles. The test results confirmed that the proposed algorithm can successfully perform evasive maneuvers on urban roads to ensure safety and mitigate collision risk with the surrounding traffic agents.
- 14 December 2021
- Published by SAGE Publications in Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering
- Vol. 236 (13) , 3005-3026
Design and implementation of human driving data-based lane-keeping assistance system for electric bus
This paper describes design, implementation, and evaluation of human driving data-based Lane Keeping Assistance System (LKAS) for electric bus equipped with a hybrid electric power steering system. The hybrid electric-power steering system used in this study means a steering system in which an Electric Power Steering (EPS) system and an Electro-Hydraulic Power Steering (EHPS) system are integrated into a ball-nut. A dynamic model of hybrid EPS system including EHPS system and EPS system has been developed to generate EPS torque and EHPS force corresponding to the input torque. In order to determine proper timing of LKAS intervention, driving data of electric bus drivers were collected and driving patterns were analyzed using a 2-D normal distribution probability density function. Lane information necessary for the lane-keeping assistance system is obtained from a vision camera mounted on the electric bus. Sliding mode control is used to get a Steering Wheel Angle (SWA) required for LKAS. A Proportional–Integral (PI) control is used to obtain an overlay torque required to track the target SWA. A proposed DLC threshold has been validated using vehicle simulation software, TruckSim, and MATLAB/Simulink. It is shown that the proposed DLC threshold shows good performance in both cases of slow lane departure and fast lane departure. The proposed algorithm has been successfully implemented on the electric bus and evaluated via real-world driving tests. Test scenario setting and the evaluation of performance were carried out by ISO 11270 criteria. It is shown that the algorithm successfully prevented the electric bus from unintended lane departure satisfying ISO 11270 criteria.
- 14 December 2021
- Published by SAGE Publications in Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering
- Vol. 236 (13) , 3005-3026
Design and implementation of human driving data-based lane-keeping assistance system for electric bus
This paper describes design, implementation, and evaluation of human driving data-based Lane Keeping Assistance System (LKAS) for electric bus equipped with a hybrid electric power steering system. The hybrid electric-power steering system used in this study means a steering system in which an Electric Power Steering (EPS) system and an Electro-Hydraulic Power Steering (EHPS) system are integrated into a ball-nut. A dynamic model of hybrid EPS system including EHPS system and EPS system has been developed to generate EPS torque and EHPS force corresponding to the input torque. In order to determine proper timing of LKAS intervention, driving data of electric bus drivers were collected and driving patterns were analyzed using a 2-D normal distribution probability density function. Lane information necessary for the lane-keeping assistance system is obtained from a vision camera mounted on the electric bus. Sliding mode control is used to get a Steering Wheel Angle (SWA) required for LKAS. A Proportional–Integral (PI) control is used to obtain an overlay torque required to track the target SWA. A proposed DLC threshold has been validated using vehicle simulation software, TruckSim, and MATLAB/Simulink. It is shown that the proposed DLC threshold shows good performance in both cases of slow lane departure and fast lane departure. The proposed algorithm has been successfully implemented on the electric bus and evaluated via real-world driving tests. Test scenario setting and the evaluation of performance were carried out by ISO 11270 criteria. It is shown that the algorithm successfully prevented the electric bus from unintended lane departure satisfying ISO 11270 criteria.
- Open Access
- 26 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 63440-63455
Interaction-Aware Probabilistic Trajectory Prediction of Cut-In Vehicles Using Gaussian Process for Proactive Control of Autonomous Vehicles
This paper presents a probabilistic trajectory prediction of cut-in vehicles exploiting the information of interacting vehicles. First, a probability distribution of behavioral parameters, which represents the characteristics of lane-change motion, is obtained via Gaussian Process Regression (GPR). For this purpose, Gaussian Process (GP) models are trained using real-world trajectories of lane-changing vehicles and adjacent vehicles. Subsequently, the future states of the lane-change vehicle are probabilistically estimated using a path-following model, which introduces virtual measurements based on the information of behavioral parameters. The proposed predictor is applied to the motion planning and control of autonomous vehicles. A Model Predictive Control (MPC) is designed to achieve predictive maneuvering of autonomous vehicles against cut-in preceding vehicles. The proposed predictor has been evaluated in terms of its prediction accuracy. Also, the performance of the proposed predictor-based control has been validated via computer simulations and autonomous driving vehicle tests. Compared to conventional prediction methods, it is shown that the interaction-aware proposed predictor provides improved prediction of cut-in vehicles' motion in multi-vehicle scenarios. Furthermore, the control results indicate that the proposed predictor helps the autonomous vehicle to reduce the control effort and improve ride quality for passengers in cut-in scenarios, while guaranteeing safety.
- 26 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 63440-63455
Open AccessInteraction-Aware Probabilistic Trajectory Prediction of Cut-In Vehicles Using Gaussian Process for Proactive Control of Autonomous Vehicles
This paper presents a probabilistic trajectory prediction of cut-in vehicles exploiting the information of interacting vehicles. First, a probability distribution of behavioral parameters, which represents the characteristics of lane-change motion, is obtained via Gaussian Process Regression (GPR). For this purpose, Gaussian Process (GP) models are trained using real-world trajectories of lane-changing vehicles and adjacent vehicles. Subsequently, the future states of the lane-change vehicle are probabilistically estimated using a path-following model, which introduces virtual measurements based on the information of behavioral parameters. The proposed predictor is applied to the motion planning and control of autonomous vehicles. A Model Predictive Control (MPC) is designed to achieve predictive maneuvering of autonomous vehicles against cut-in preceding vehicles. The proposed predictor has been evaluated in terms of its prediction accuracy. Also, the performance of the proposed predictor-based control has been validated via computer simulations and autonomous driving vehicle tests. Compared to conventional prediction methods, it is shown that the interaction-aware proposed predictor provides improved prediction of cut-in vehicles' motion in multi-vehicle scenarios. Furthermore, the control results indicate that the proposed predictor helps the autonomous vehicle to reduce the control effort and improve ride quality for passengers in cut-in scenarios, while guaranteeing safety.