Playing Flappy Bird Based on Motion Recognition Using a Transformer Model and LIDAR Sensor
A transformer neural network is employed in the present study to predict Q-values in a simulated environment using reinforcement learning techniques. The goal is to teach an agent to navigate and excel in the Flappy Bird game, which became a popular model for control in machine learning approaches....
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/24/6/1905 |
| _version_ | 1827305061754077184 |
|---|---|
| author | Iveta Dirgová Luptáková Martin Kubovčík Jiří Pospíchal |
| author_facet | Iveta Dirgová Luptáková Martin Kubovčík Jiří Pospíchal |
| author_sort | Iveta Dirgová Luptáková |
| collection | DOAJ |
| description | A transformer neural network is employed in the present study to predict Q-values in a simulated environment using reinforcement learning techniques. The goal is to teach an agent to navigate and excel in the Flappy Bird game, which became a popular model for control in machine learning approaches. Unlike most top existing approaches that use the game’s rendered image as input, our main contribution lies in using sensory input from LIDAR, which is represented by the ray casting method. Specifically, we focus on understanding the temporal context of measurements from a ray casting perspective and optimizing potentially risky behavior by considering the degree of the approach to objects identified as obstacles. The agent learned to use the measurements from ray casting to avoid collisions with obstacles. Our model substantially outperforms related approaches. Going forward, we aim to apply this approach in real-world scenarios. |
| first_indexed | 2024-04-24T17:49:41Z |
| format | Article |
| id | doaj.art-a4702b01e2964a6fbe599e3b9b1e4ef5 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-24T17:49:41Z |
| publishDate | 2024-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-a4702b01e2964a6fbe599e3b9b1e4ef52024-03-27T14:04:06ZengMDPI AGSensors1424-82202024-03-01246190510.3390/s24061905Playing Flappy Bird Based on Motion Recognition Using a Transformer Model and LIDAR SensorIveta Dirgová Luptáková0Martin Kubovčík1Jiří Pospíchal2Institute of Computer Technologies and Informatics, Faculty of Natural Sciences, University of Ss. Cyril and Methodius, J. Herdu 2, 917 01 Trnava, SlovakiaInstitute of Computer Technologies and Informatics, Faculty of Natural Sciences, University of Ss. Cyril and Methodius, J. Herdu 2, 917 01 Trnava, SlovakiaInstitute of Computer Technologies and Informatics, Faculty of Natural Sciences, University of Ss. Cyril and Methodius, J. Herdu 2, 917 01 Trnava, SlovakiaA transformer neural network is employed in the present study to predict Q-values in a simulated environment using reinforcement learning techniques. The goal is to teach an agent to navigate and excel in the Flappy Bird game, which became a popular model for control in machine learning approaches. Unlike most top existing approaches that use the game’s rendered image as input, our main contribution lies in using sensory input from LIDAR, which is represented by the ray casting method. Specifically, we focus on understanding the temporal context of measurements from a ray casting perspective and optimizing potentially risky behavior by considering the degree of the approach to objects identified as obstacles. The agent learned to use the measurements from ray casting to avoid collisions with obstacles. Our model substantially outperforms related approaches. Going forward, we aim to apply this approach in real-world scenarios.https://www.mdpi.com/1424-8220/24/6/1905reinforcement learningmotion sensorsray castingsignal processingtime series processingtransformer model |
| spellingShingle | Iveta Dirgová Luptáková Martin Kubovčík Jiří Pospíchal Playing Flappy Bird Based on Motion Recognition Using a Transformer Model and LIDAR Sensor Sensors reinforcement learning motion sensors ray casting signal processing time series processing transformer model |
| title | Playing Flappy Bird Based on Motion Recognition Using a Transformer Model and LIDAR Sensor |
| title_full | Playing Flappy Bird Based on Motion Recognition Using a Transformer Model and LIDAR Sensor |
| title_fullStr | Playing Flappy Bird Based on Motion Recognition Using a Transformer Model and LIDAR Sensor |
| title_full_unstemmed | Playing Flappy Bird Based on Motion Recognition Using a Transformer Model and LIDAR Sensor |
| title_short | Playing Flappy Bird Based on Motion Recognition Using a Transformer Model and LIDAR Sensor |
| title_sort | playing flappy bird based on motion recognition using a transformer model and lidar sensor |
| topic | reinforcement learning motion sensors ray casting signal processing time series processing transformer model |
| url | https://www.mdpi.com/1424-8220/24/6/1905 |
| work_keys_str_mv | AT ivetadirgovaluptakova playingflappybirdbasedonmotionrecognitionusingatransformermodelandlidarsensor AT martinkubovcik playingflappybirdbasedonmotionrecognitionusingatransformermodelandlidarsensor AT jiripospichal playingflappybirdbasedonmotionrecognitionusingatransformermodelandlidarsensor |