• 전기전자학회논문지
• /
• 제23권4호
• /
• pp.1150-1156
• /
• 2019

This paper explores a model-free value-based approach for solving survival gridworld problem. Survival gridworld problem opens up a challenge involving taking risks to gain better rewards. Classic value-based approach in model-free reinforcement learning assumes minimal risk decisions. The proposed method involves a hybrid on-policy and off-policy updates to experience roll-outs using a modified Q-based update equation that introduces a parametric linear rectifier and motivational discount. The significance of this approach is it allows model-free training of agents that take into account risk factors and motivated exploration to gain better path decisions. Experimentations suggest that the proposed method achieved better exploration and path selection resulting to higher episode scores than classic off-policy and on-policy Q-based updates.

• 한국군사과학기술학회지
• /
• 제23권4호
• /
• pp.399-406
• /
• 2020

In this paper, we propose a time-varying proportional navigation guidance law that determines the proportional navigation gain in real-time according to the operating situation. When intercepting a target, an unidentified evasion strategy causes a loss of optimality. To compensate for this problem, proper proportional navigation gain is derived at every time step by solving an optimal control problem with the inferred evader's strategy. Recently, deep reinforcement learning algorithms are introduced to deal with complex optimal control problem efficiently. We adapt the actor-critic method to build a proportional navigation gain network and the network is trained by the Proximal Policy Optimization(PPO) algorithm to learn an evasion strategy of the target. Numerical experiments show the effectiveness and optimality of the proposed method.

• 로봇학회논문지
• /
• 제18권2호
• /
• pp.143-154
• /
• 2023

In this study, a fast motion planning method for the swing motion of a 6x6 wheel-legged robot to traverse large obstacles and gaps is proposed. The motion planning method presented in the previous paper, which was based on trajectory optimization, took up to tens of seconds and was limited to two-dimensional, structured vertical obstacles and trenches. A deep neural network based on one-dimensional Convolutional Neural Network (CNN) is introduced to generate keyframes, which are then used to represent smooth reference commands for the six leg angles along the robot's path. The network is initially trained using the behavioral cloning method with a dataset gathered from previous simulation results of the trajectory optimization. Its performance is then improved through reinforcement learning, using a one-step REINFORCE algorithm. The trained model has increased the speed of motion planning by up to 820 times and improved the success rates of obstacle crossing under harsh conditions, such as low friction and high roughness.

• 한국인터넷방송통신학회논문지
• /
• 제24권3호
• /
• pp.15-20
• /
• 2024

무인 항공기의 경로 계획은 고정 및 동적 장애물을 포함하는 복합 환경에서 장애물 충돌을 회피하는 것이 중요하다. RRT나 A^*와 같은 경로 계획 알고리즘은 고정된 장애물 회피를 효과적으로 수행하지만, 고차원 환경일수록 계산 복잡도가 증가하는 한계점을 가진다. 강화학습 기반 알고리즘은 복합적인 환경 반영이 가능하지만, 기존 경로 계획 알고리즘과 같이 고차원 환경일수록 훈련 복잡도가 증가하여 수렴성을 기대하기 힘들다. 본 논문은 셀 분해 알고리즘을 활용한 강화학습 모델을 제안한다. 제안한 모델은 학습 환경을 세부적으로 분해하여 환경의 복잡도를 감소시킨다. 또한, 에이전트의 유효한 행동을 설정하여 장애물 회피 성능을 개선한다. 이를 통해 강화학습의 탐험 문제를 해결하고, 학습의 수렴성을 높인다. 시뮬레이션 결과는 제안된 모델이 일반적인 환경의 강화학습 모델과 비교하여 학습 속도를 개선하고 효율적인 경로를 계획할 수 있음을 보여준다.

• 한국공간구조학회논문집
• /
• 제23권2호
• /
• pp.29-36
• /
• 2023

Reinforcement learning (RL) is widely applied to various engineering fields. Especially, RL has shown successful performance for control problems, such as vehicles, robotics, and active structural control system. However, little research on application of RL to optimal structural design has conducted to date. In this study, the possibility of application of RL to structural design of reinforced concrete (RC) beam was investigated. The example of RC beam structural design problem introduced in previous study was used for comparative study. Deep q-network (DQN) is a famous RL algorithm presenting good performance in the discrete action space and thus it was used in this study. The action of DQN agent is required to represent design variables of RC beam. However, the number of design variables of RC beam is too many to represent by the action of conventional DQN. To solve this problem, multi-agent DQN was used in this study. For more effective reinforcement learning process, DDQN (Double Q-Learning) that is an advanced version of a conventional DQN was employed. The multi-agent of DDQN was trained for optimal structural design of RC beam to satisfy American Concrete Institute (318) without any hand-labeled dataset. Five agents of DDQN provides actions for beam with, beam depth, main rebar size, number of main rebar, and shear stirrup size, respectively. Five agents of DDQN were trained for 10,000 episodes and the performance of the multi-agent of DDQN was evaluated with 100 test design cases. This study shows that the multi-agent DDQN algorithm can provide successfully structural design results of RC beam.

• 한국컴퓨터그래픽스학회논문지
• /
• 제26권4호
• /
• pp.1-8
• /
• 2020

캐릭터로 하여금 시뮬레이션 내에서 사용자가 원하는 동작을 보이도록 물리 기반 제어기를 만들 수 있다면 주변 환경의 변화와 다른 캐릭터와의 상호작용에 대하여 자연스러운 반응을 보이는 캐릭터 애니메이션을 생성할 수 있다. 최근 심층 강화 학습을 이용해 물리 기반 제어기가 더 안정적이고 다양한 동작을 합성하도록 하는 연구가 다수 이루어져 왔다. 본 논문에서는 다리가 하나 달린 픽사 애니메이션 스튜디오의 마스코트 캐릭터 Luxo를 주어진 목적지까지 뛰어 도착하게 하는 심층 강화학습 모델을 제시한다. 효율적으로 뛰는 동작을 학습하도록 하기 위해서 Luxo의 각 관절의 각도값들을 선형 보간법으로 생성하여 참조 모션을 만들었으며, 캐릭터는 이를 모방하면서 균형을 유지하여 목표한 위치까지 도달하도록 하는 제어 정책(control policy)을 학습한다. 참조 동작을 사용하지 않고 Luxo 동작을 제어하도록 학습된 정책과 비교한 실험 결과, 제안된 방법을 사용하면 사용자가 지정한 위치로 Luxo가 점프하며 이동하는 정책을 더 효율적으로 학습할 수 있었다.

• 한국인터넷방송통신학회논문지
• /
• 제23권6호
• /
• pp.33-39
• /
• 2023

최근 비행 애드-훅 네트워크(Flying Ad-hoc Network) 환경에서 강화학습을 이용한 통신 성능 개선과 이동성 모델 설계에 관한 연구가 진행되고 있다. 무인항공기(UAV)에서의 이동성 모델은 움직임을 예측하고 제어하기 위한 핵심요소로 주목받고 있다. 본 논문에서는 무인항공기가 운용되는 3차원 가상 환경을 구현하고, 무인항공기의 경로 최적화를 위해 푸리에 기저 함수 근사를 적용한 Q-learning과 DQN 두 가지 강화학습 알고리즘을 적용하여 모델을 설계 및 성능을 분석하였다. 실험 결과를 통해 3차원 가상 환경에서 DQN 모델이 Q-learning 모델 대비 최적의 경로 탐색에 적합한 것을 확인하였다.

• 한국산업융합학회 논문집
• /
• 제25권3호
• /
• pp.467-476
• /
• 2022

The Bank of Korea raised the benchmark interest rate by a quarter percentage point to 1.75 percent per year, and analysts predict that South Korea's policy rate will reach 2.00 percent by the end of calendar year 2022. Furthermore, because market volatility has been significantly increased by a variety of factors, including rising rates, inflation, and market volatility, many investors have struggled to meet their financial objectives or deliver returns. Banks and financial institutions are attempting to provide Robo-Advisors to manage client portfolios without human intervention in this situation. In this regard, determining the best hyper-parameter combination is becoming increasingly important. This study compares some activation functions of the Deep Deterministic Policy Gradient(DDPG) and Twin-delayed Deep Deterministic Policy Gradient (TD3) Algorithms to choose a sequence of actions that maximizes long-term reward. The DDPG and TD3 outperformed its benchmark index, according to the results. One reason for this is that we need to understand the action probabilities in order to choose an action and receive a reward, which we then compare to the state value to determine an advantage. As interest in machine learning has grown and research into deep reinforcement learning has become more active, finding an optimal hyper-parameter combination for DDPG and TD3 has become increasingly important.

• 인터넷정보학회논문지
• /
• 제24권2호
• /
• pp.1-10
• /
• 2023

The Multi-access Edge Computing (MEC) paradigm equips network edge telecommunication infrastructure with cloud computing resources. It seeks to transform the edge into an IT services platform for hosting resource-intensive and delay-stringent services for mobile users, thereby significantly enhancing perceived service quality of experience. However, erratic user mobility impedes seamless service continuity as well as satisfying delay-stringent service requirements, especially as users roam farther away from the serving MEC resource, which deteriorates quality of experience. This work proposes a deep reinforcement learning based service mobility management approach for ensuring seamless migration of service instances along user mobility. The proposed approach focuses on the problem of selecting the optimal MEC resource to host services for high mobility users, thereby reducing service migration rejection rate and enhancing service availability. Efficacy of the proposed approach is confirmed through simulation experiments, where results show that on average, the proposed scheme reduces service delay by 8%, task computing time by 36%, and migration rejection rate by more than 90%, when comparing to a baseline scheme.

• 한국건설관리학회논문집
• /
• 제24권5호
• /
• pp.73-82
• /
• 2023

Optimization of Construction Site Layout Planning (CSLP) heavily relies on workers' travel paths. However, traditional path generation approaches predominantly focus on the shortest path, often neglecting critical variables such as individual wayfinding tendencies, the spatial arrangement of site objects, and potential hazards. These oversights can lead to compromised path simulations, resulting in less reliable site layout plans. While Deep Reinforcement Learning (DRL) has been proposed as a potential alternative to address these issues, it has shown limitations. Despite presenting more realistic travel paths by considering these variables, DRL often struggles with efficiency in complex environments, leading to extended learning times and potential failures. To overcome these challenges, this study introduces a refined model that enhances spatial navigation capabilities and learning performance by integrating workers' visibility into the reward functions. The proposed model demonstrated a 12.47% increase in the pathfinding success rate and notable improvements in the other two performance measures compared to the existing DRL framework. The adoption of this model could greatly enhance the reliability of the results, ultimately improving site operational efficiency and safety management such as by reducing site congestion and accidents. Future research could expand this study by simulating travel paths in dynamic, multi-agent environments that represent different stages of construction.