LiDAR DNN based self-attitude estimation with learning landscape regularities
| Autor: | Ryota Ozaki, Naoya Sugiura, Yoji Kuroda |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
Attitude estimation
Mobile robotics Deep learning Extended Kalman filter Technology Mechanical engineering and machinery TJ1-1570 Control engineering systems. Automatic machinery (General) TJ212-225 Machine design and drawing TJ227-240 Technology (General) T1-995 Industrial engineering. Management engineering T55.4-60.8 Automation T59.5 Information technology T58.5-58.64 |
| Zdroj: | ROBOMECH Journal, Vol 8, Iss 1, Pp 1-12 (2021) |
| Druh dokumentu: | article |
| ISSN: | 2197-4225 |
| DOI: | 10.1186/s40648-021-00213-5 |
| Popis: | Abstract This paper presents an EKF (extended Kalman filter) based self-attitude estimation method with a LiDAR DNN (deep neural network) learning landscape regularities. The proposed DNN infers the gravity direction from LiDAR data. The point cloud obtained with the LiDAR is transformed to a depth image to be input to the network. It is pre-trained with large synthetic datasets. They are collected in a flight simulator because various gravity vectors can be easily obtained, although this study focuses not only on UAVs. Fine-tuning with datasets collected with real sensors is done after the pre-training. Data augmentation is processed during the training in order to provide higher general versatility. The proposed method integrates angular rates from a gyroscope and the DNN outputs in an EKF. Static validations are performed to show the DNN can infer the gravity direction. Dynamic validations are performed to show the DNN can be used in real-time estimation. Some conventional methods are implemented for comparison. |
| Databáze: | Directory of Open Access Journals |
| Externí odkaz: |
|
Full text from SpringerLink