Autonomous Navigation via Deep Reinforcement Learning for Resource Constraint Edge Nodes Using Transfer Learning
Smart and agile drones are fast becoming ubiquitous at the edge of the cloud. The usage of these drones is constrained by their limited power and compute capability. In this paper, we present a Transfer Learning (TL) based approach to reduce on-board computation required to train a deep neural netwo...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8978577/ |
| _version_ | 1819171336755347456 |
|---|---|
| author | Aqeel Anwar Arijit Raychowdhury |
| author_facet | Aqeel Anwar Arijit Raychowdhury |
| author_sort | Aqeel Anwar |
| collection | DOAJ |
| description | Smart and agile drones are fast becoming ubiquitous at the edge of the cloud. The usage of these drones is constrained by their limited power and compute capability. In this paper, we present a Transfer Learning (TL) based approach to reduce on-board computation required to train a deep neural network for autonomous navigation via value-based Deep Reinforcement Learning for a target algorithmic performance. A library of 3D realistic meta-environments is manually designed using Unreal Gaming Engine and the network is trained end-to-end. These trained meta-weights are then used as initializers to the network in a test environment and fine-tuned for the last few fully connected layers. Variation in drone dynamics and environmental characteristics is carried out to show robustness of the approach. Using NVIDIA GPU profiler, it was shown that the energy consumption and training latency is reduced by 3.7$\times$ and 1.8$\times$ respectively without significant degradation in the performance in terms of average distance traveled before crash i.e. Mean Safe Flight (MSF). The approach is also tested on a real environment using DJI Tello drone and similar results were reported. The code for the approach can be found on GitHub: https://github.com/aqeelanwar/DRLwithTL. |
| first_indexed | 2024-12-22T19:49:40Z |
| format | Article |
| id | doaj.art-e1c6dbd5a18645e391c5d682f4b78e79 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-22T19:49:40Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-e1c6dbd5a18645e391c5d682f4b78e792022-12-21T18:14:35ZengIEEEIEEE Access2169-35362020-01-018265492656010.1109/ACCESS.2020.29711728978577Autonomous Navigation via Deep Reinforcement Learning for Resource Constraint Edge Nodes Using Transfer LearningAqeel Anwar0https://orcid.org/0000-0001-6768-058XArijit Raychowdhury1https://orcid.org/0000-0001-8391-0576Department of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USADepartment of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USASmart and agile drones are fast becoming ubiquitous at the edge of the cloud. The usage of these drones is constrained by their limited power and compute capability. In this paper, we present a Transfer Learning (TL) based approach to reduce on-board computation required to train a deep neural network for autonomous navigation via value-based Deep Reinforcement Learning for a target algorithmic performance. A library of 3D realistic meta-environments is manually designed using Unreal Gaming Engine and the network is trained end-to-end. These trained meta-weights are then used as initializers to the network in a test environment and fine-tuned for the last few fully connected layers. Variation in drone dynamics and environmental characteristics is carried out to show robustness of the approach. Using NVIDIA GPU profiler, it was shown that the energy consumption and training latency is reduced by 3.7$\times$ and 1.8$\times$ respectively without significant degradation in the performance in terms of average distance traveled before crash i.e. Mean Safe Flight (MSF). The approach is also tested on a real environment using DJI Tello drone and similar results were reported. The code for the approach can be found on GitHub: https://github.com/aqeelanwar/DRLwithTL.https://ieeexplore.ieee.org/document/8978577/Autonomous navigationtransfer learningdeep reinforcement learningdrone |
| spellingShingle | Aqeel Anwar Arijit Raychowdhury Autonomous Navigation via Deep Reinforcement Learning for Resource Constraint Edge Nodes Using Transfer Learning IEEE Access Autonomous navigation transfer learning deep reinforcement learning drone |
| title | Autonomous Navigation via Deep Reinforcement Learning for Resource Constraint Edge Nodes Using Transfer Learning |
| title_full | Autonomous Navigation via Deep Reinforcement Learning for Resource Constraint Edge Nodes Using Transfer Learning |
| title_fullStr | Autonomous Navigation via Deep Reinforcement Learning for Resource Constraint Edge Nodes Using Transfer Learning |
| title_full_unstemmed | Autonomous Navigation via Deep Reinforcement Learning for Resource Constraint Edge Nodes Using Transfer Learning |
| title_short | Autonomous Navigation via Deep Reinforcement Learning for Resource Constraint Edge Nodes Using Transfer Learning |
| title_sort | autonomous navigation via deep reinforcement learning for resource constraint edge nodes using transfer learning |
| topic | Autonomous navigation transfer learning deep reinforcement learning drone |
| url | https://ieeexplore.ieee.org/document/8978577/ |
| work_keys_str_mv | AT aqeelanwar autonomousnavigationviadeepreinforcementlearningforresourceconstraintedgenodesusingtransferlearning AT arijitraychowdhury autonomousnavigationviadeepreinforcementlearningforresourceconstraintedgenodesusingtransferlearning |