Toward Self-Driving Bicycles Using State-of-the-Art Deep Reinforcement Learning Algorithms
In this paper, we propose a controller for a bicycle using the DDPG (Deep Deterministic Policy Gradient) algorithm, which is a state-of-the-art deep reinforcement learning algorithm. We use a reward function and a deep neural network to build the controller. By using the proposed controller, a bicyc...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-02-01
|
| Series: | Symmetry |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-8994/11/2/290 |
| _version_ | 1811299680009584640 |
|---|---|
| author | SeungYoon Choi Tuyen P. Le Quang D. Nguyen Md Abu Layek SeungGwan Lee TaeChoong Chung |
| author_facet | SeungYoon Choi Tuyen P. Le Quang D. Nguyen Md Abu Layek SeungGwan Lee TaeChoong Chung |
| author_sort | SeungYoon Choi |
| collection | DOAJ |
| description | In this paper, we propose a controller for a bicycle using the DDPG (Deep Deterministic Policy Gradient) algorithm, which is a state-of-the-art deep reinforcement learning algorithm. We use a reward function and a deep neural network to build the controller. By using the proposed controller, a bicycle can not only be stably balanced but also travel to any specified location. We confirm that the controller with DDPG shows better performance than the other baselines such as Normalized Advantage Function (NAF) and Proximal Policy Optimization (PPO). For the performance evaluation, we implemented the proposed algorithm in various settings such as fixed and random speed, start location, and destination location. |
| first_indexed | 2024-04-13T06:39:29Z |
| format | Article |
| id | doaj.art-8b969cc31ac149ab82a460dd86f92968 |
| institution | Directory Open Access Journal |
| issn | 2073-8994 |
| language | English |
| last_indexed | 2024-04-13T06:39:29Z |
| publishDate | 2019-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Symmetry |
| spelling | doaj.art-8b969cc31ac149ab82a460dd86f929682022-12-22T02:57:47ZengMDPI AGSymmetry2073-89942019-02-0111229010.3390/sym11020290sym11020290Toward Self-Driving Bicycles Using State-of-the-Art Deep Reinforcement Learning AlgorithmsSeungYoon Choi0Tuyen P. Le1Quang D. Nguyen2Md Abu Layek3SeungGwan Lee4TaeChoong Chung5Artificial Intelligence Lab, Computer Science and Engineering, Kyung Hee University, Yongin-si, Gyonggi-do, Gyeonggi 446-701, KoreaArtificial Intelligence Lab, Computer Science and Engineering, Kyung Hee University, Yongin-si, Gyonggi-do, Gyeonggi 446-701, KoreaArtificial Intelligence Lab, Computer Science and Engineering, Kyung Hee University, Yongin-si, Gyonggi-do, Gyeonggi 446-701, KoreaArtificial Intelligence Lab, Computer Science and Engineering, Kyung Hee University, Yongin-si, Gyonggi-do, Gyeonggi 446-701, KoreaHumanitas College, Kyung Hee University, Yongin, Gyeonggi 446-701, KoreaArtificial Intelligence Lab, Computer Science and Engineering, Kyung Hee University, Yongin-si, Gyonggi-do, Gyeonggi 446-701, KoreaIn this paper, we propose a controller for a bicycle using the DDPG (Deep Deterministic Policy Gradient) algorithm, which is a state-of-the-art deep reinforcement learning algorithm. We use a reward function and a deep neural network to build the controller. By using the proposed controller, a bicycle can not only be stably balanced but also travel to any specified location. We confirm that the controller with DDPG shows better performance than the other baselines such as Normalized Advantage Function (NAF) and Proximal Policy Optimization (PPO). For the performance evaluation, we implemented the proposed algorithm in various settings such as fixed and random speed, start location, and destination location.https://www.mdpi.com/2073-8994/11/2/290deep reinforcement learningdeep deterministic policy gradient (DDPG)machine learningself-driving bicycle |
| spellingShingle | SeungYoon Choi Tuyen P. Le Quang D. Nguyen Md Abu Layek SeungGwan Lee TaeChoong Chung Toward Self-Driving Bicycles Using State-of-the-Art Deep Reinforcement Learning Algorithms Symmetry deep reinforcement learning deep deterministic policy gradient (DDPG) machine learning self-driving bicycle |
| title | Toward Self-Driving Bicycles Using State-of-the-Art Deep Reinforcement Learning Algorithms |
| title_full | Toward Self-Driving Bicycles Using State-of-the-Art Deep Reinforcement Learning Algorithms |
| title_fullStr | Toward Self-Driving Bicycles Using State-of-the-Art Deep Reinforcement Learning Algorithms |
| title_full_unstemmed | Toward Self-Driving Bicycles Using State-of-the-Art Deep Reinforcement Learning Algorithms |
| title_short | Toward Self-Driving Bicycles Using State-of-the-Art Deep Reinforcement Learning Algorithms |
| title_sort | toward self driving bicycles using state of the art deep reinforcement learning algorithms |
| topic | deep reinforcement learning deep deterministic policy gradient (DDPG) machine learning self-driving bicycle |
| url | https://www.mdpi.com/2073-8994/11/2/290 |
| work_keys_str_mv | AT seungyoonchoi towardselfdrivingbicyclesusingstateoftheartdeepreinforcementlearningalgorithms AT tuyenple towardselfdrivingbicyclesusingstateoftheartdeepreinforcementlearningalgorithms AT quangdnguyen towardselfdrivingbicyclesusingstateoftheartdeepreinforcementlearningalgorithms AT mdabulayek towardselfdrivingbicyclesusingstateoftheartdeepreinforcementlearningalgorithms AT seunggwanlee towardselfdrivingbicyclesusingstateoftheartdeepreinforcementlearningalgorithms AT taechoongchung towardselfdrivingbicyclesusingstateoftheartdeepreinforcementlearningalgorithms |