Heoncheol Lee’s research while affiliated with Kumoh National Institute of Technology and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (16)


Figure 1. Overall flowchart of the Dueling-DQN-based routing algorithm.
Routing algorithm research related to reinforcement learning.
Algorithm's hardware resource usage.
An FPGA-Accelerated CNN with Parallelized Sum Pooling for Onboard Realtime Routing in Dynamic Low-Orbit Satellite Networks
  • Article
  • Full-text available

June 2024

·

11 Reads

·

1 Citation

Electronics

Hyeonwoo Kim

·

Juhyeon Park

·

Heoncheol Lee

·

[...]

·

Myonghun Han

This paper addresses the problem of real-time onboard routing for dynamic low earth orbit (LEO) satellite networks. It is difficult to apply general routing algorithms to dynamic LEO networks due to the frequent changes in satellite topology caused by the disconnection between moving satellites. Deep reinforcement learning (DRL) models trained by various dynamic networks can be considered. However, since the inference process with the DRL model requires too long a computation time due to multiple convolutional layer operations, it is not practical to apply to a real-time on-board computer (OBC) with limited computing resources. To solve the problem, this paper proposes a practical co-design method with heterogeneous processors to parallelize and accelerate a part of the multiple convolutional layer operations on a field-programmable gate array (FPGA). The proposed method was tested with a real heterogeneous processor-based OBC and showed that the proposed method was about 3.10 times faster than the conventional method while achieving the same routing results.

Download

FPGA-based Inference Parallelization for Onboard RL-based Routing in Dynamic LEO Satellite Networks

April 2024

·

11 Reads

International Journal of Aeronautical and Space Sciences

This paper addresses the problem of onboard computer application of dynamic low-orbit satellite network routing algorithms. In low-orbit satellite networks, the satellite topology changes in real time, and satellite disconnection occurs frequently. The problem of routing algorithms for low-orbit satellites can be solved by reinforcement learning algorithms. However, the inference process based on deep reinforcement learning models suffers from excessive computation due to the operation of multiple convolutional layers. In this paper, we propose a method to accelerate convolutional layer operations by parallelizing them using heterogeneous processors. This approach is compared to the traditional single-processor-based convolutional operation method, commonly used in dynamic low-orbit satellite network routing algorithms. Our evaluation, conducted on an actual heterogeneous processor-based onboard computer, demonstrates that the proposed method not only matches the accuracy of the conventional single-processor-based approach, but also significantly reduces the execution time.


Parallelized Particle Swarm Optimization on FPGA for Realtime Ballistic Target Tracking

October 2023

·

40 Reads

·

2 Citations

Sensors

This paper addresses the problem of tracking a high-speed ballistic target in real time. Particle swarm optimization (PSO) can be a solution to overcome the motion of the ballistic target and the nonlinearity of the measurement model. However, in general, particle swarm optimization requires a great deal of computation time, so it is difficult to apply to realtime systems. In this paper, we propose a parallelized particle swarm optimization technique using field-programmable gate array (FPGA) to be accelerated for realtime ballistic target tracking. The realtime performance of the proposed method has been tested and analyzed on a well-known heterogeneous processing system with a field-programmable gate array. The proposed parallelized particle swarm optimization was successfully conducted on the heterogeneous processing system and produced similar tracking results. Also, compared to conventional particle swarm optimization, which is based on the only central processing unit, the computation time is significantly reduced by up to 3.89×.








Figure 2. The flowchart of the proposed method.
Figure 5. Simulation results of the nearest neighbor in virtual space for: (a) 20 nodes, (b) 30 nodes, (c) 40 nodes. The red dot is the starting point and the blue dots are the nodes to visit. Each color line is the path of each robot.
Figure 9. Simulation results of the nearest neighbor in Antarctic environments for: (a) 10 nodes, (b) 20 nodes, (c) 30 nodes. The red dot is the starting point and the blue dots are the nodes to visit. Each color line is the path of each robot.
The elevation distance comparison results of the simulation.
The elevation distance comparison results of real Antarctic environments.
Multi-Robot Task Scheduling with Ant Colony Optimization in Antarctic Environments

January 2023

·

50 Reads

·

6 Citations

Sensors

This paper addresses the problem of multi-robot task scheduling in Antarctic environments. There are various algorithms for multi-robot task scheduling, but there is a risk in robot operation when applied in Antarctic environments. This paper proposes a practical multi-robot scheduling method using ant colony optimization in Antarctic environments. The proposed method was tested in both simulated and real Antarctic environments, and it was analyzed and compared with other existing algorithms. The improved performance of the proposed method was verified by finding more efficiently scheduled multiple paths with lower costs than the other algorithms.


Citations (9)


... In [67], the authors address the problem of accelerating Deep reinforcement learning (DRL) models onboard satellites. The application addressed concerns about the onboard real-time routing for dynamic low Earth orbit (LEO) satellite networks. ...

Reference:

Review on Hardware Devices and Software Techniques Enabling Neural Network Inference Onboard Satellites
An FPGA-Accelerated CNN with Parallelized Sum Pooling for Onboard Realtime Routing in Dynamic Low-Orbit Satellite Networks

Electronics

... The computational efficiency of the proposed method is relatively low because it uses PSO for its parameter estimation. Recently, parallel PSO calculations using GPUs [27] have been proposed, and it is expected that their calculation speed will be improved. In the data analysis of target 2, it was confirmed that respiration harmonics affect the parameter estimation of the heartbeat. ...

Parallelized Particle Swarm Optimization on FPGA for Realtime Ballistic Target Tracking

Sensors

... To address this, a probabilistic model using a Gaussian distribution is applied to crevasse data [26], considering the variability in crevasse positions and sizes. Previous studies have proposed methods to minimize steep slopes while visiting destinations, but they were limited in fully reflecting the Antarctic environment [27,28]. Therefore, this study improves scheduling by integrating crevasse data with the elevation information used in previous research to find safer routes. ...

Efficient Multi-task Scheduling with Nearest Neighbor Algorithm for Multi-robot Systems in Antarctic Environments
  • Citing Article
  • April 2023

Journal of Institute of Control Robotics and Systems

... The challenges posed by these factors are addressed, and effective solutions to enhance the efficiency and accuracy of ballistic target tracking using particle filters are proposed in this paper. Most research has largely focused on the acceleration of Particle Filters using Graphics Processing Units (GPUs) [7], [8], [9]. Significant results in terms of increased computational speed have been yielded by this approach, but it causes the problem of power consumption [10]. ...

Accelerated Particle Filter With GPU for Real-Time Ballistic Target Tracking

IEEE Access

... To address this, a probabilistic model using a Gaussian distribution is applied to crevasse data [26], considering the variability in crevasse positions and sizes. Previous studies have proposed methods to minimize steep slopes while visiting destinations, but they were limited in fully reflecting the Antarctic environment [27,28]. Therefore, this study improves scheduling by integrating crevasse data with the elevation information used in previous research to find safer routes. ...

Multi-Robot Task Scheduling with Ant Colony Optimization in Antarctic Environments

Sensors

... Given the high reliability requirements of designing missile control systems, traditional missile control system designs largely incorporate fixed supplementary and feedback techniques, which either control the missile system through frequency or provide structural analysis within the feedback frequency loop. With the development of theory, literature has shown that the concept of missile control system, which combines modern control theory and methods, gradually permeates into various aspects of missile autopilot design, guidance mode design, and so on (Lee et al. 2022). In order to comprehensively analyze the aviation capabilities of aviation guided missiles, such as pole configuration, variable structure control, and so on. ...

GPU-Accelerated PD-IPM for Real-Time Model Predictive Control in Integrated Missile Guidance and Control Systems

Sensors

... A PRM [29,30] is a multiple-query sampling method that is an entire-area route-planning method for generating the optimal route. A PRM generates a route efficiently within a short time in the complex and high-dimensional space, and it is applied in various sectors including robot engineering and autonomous driving [31]. The PRM generated a roadmap that connected the nodes without collisions, which are randomly selected from a certain starting point. ...

Directionally-Exploring Random Trees for Efficient Robot Path Planning in Corridor Environments
  • Citing Article
  • April 2022

The Journal of Korean Institute of Information Technology

... However, this work only focuses on the people and neglects other dynamic objects. Based on the LeGO-LOAM, Kim et al. [46] adopted YOLOv3 to multi-target recognition and described the position of targets on the reconstructed indoor 3D map. Wu et al. [47] proposed a lightweight network version of YOLOv3 by replacing the original darknet-53 backbone network with darknet-19, achieving high-speed detection. ...

Implementation of a Mobile Multi-Target Search System with 3D SLAM and Object Localization in Indoor Environments
  • Citing Conference Paper
  • October 2021

... The indirect map merging acquires the map transformation matrix by finding and matching the overlapping areas of the individual maps of robots, which is called map matching. The detailed categorization of them and the brief descriptions of the previous works are summarized in [7,8]. They have their own advantages, but they require commonly an optimization method to update the MTM more accurately regardless of the type of map merging. ...

Selective spectral correlation for efficient map merging in multi‐robot systems