Name
Abstract | ||
|---|---|---|
The paper investigates the trajectory tracking problem of a flight vehicle performing complex maneuvers. A learnable Extended Kalman Filtering (L-EKF) method is proposed. First, two recurrent neural networks, named Input Modification Network and Gain Modification Network, are designed to identify the model inaccuracy and compensate for the estimation error of the EKF. Then, the L-EKF algorithm is proposed by embedding the two networks into the EKF algorithm. Then, the proposed L-EKF method is applied to a trajectory tracking problem of a gliding vehicle with complex maneuvers. The simulation results show that the proposed method has a higher estimation accuracy and better dynamic performance than the EKF and Adaptive EKF method. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.23919/ECC.2019.8796088 | 2019 18TH EUROPEAN CONTROL CONFERENCE (ECC) |
Field | DocType | Citations |
Extended Kalman filter,Embedding,Computer science,Algorithm,Recurrent neural network,Kalman filter,Ekf algorithm,Trajectory | Conference | 0 |
PageRank | References | Authors |
0.34 | 0 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Tianyu Zheng | 1 | 0 | 0.34 |
| Yao Yu | 2 | 78 | 22.67 |
| Fenghua He | 3 | 68 | 11.36 |
| Xinran Zhang | 4 | 38 | 12.02 |