Kerui Jin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Kerui Jin has a total of 3 co-authors. Here are the top co-authors by number of shared publications:
- Guanyu Zou
School of Energy and Power Engineering, Nanjing University of Science and Technology (NJUST), Nanjing 210094, China
1 shared publication - Hongyang Bai
Nanjing University of Science and Technology
1 shared publication - Yu Cao
School of Energy and Power Engineering, Nanjing University of Science and Technology (NJUST), Nanjing 210094, China
1 shared publication
Latest Publications
- Open Access
- 2 July 2023
- Published by MDPI AG in Electronics
- Vol. 12 (13) , 2905
An GNSS/INS Integrated Navigation Algorithm Based on PSO-LSTM in Satellite Rejection
When the satellite signal is lost or interfered with, the traditional GNSS (Global Navigation Satellite System)/INS (Inertial Navigation System) integrated navigation will degenerate into INS, which results in the decrease in navigation accuracy. To solve these problems, this paper mainly established the PSO (particle swarm optimization) -LSTM (Long Short-Term Memory) neural network model to predict the increment of GNSS position under the condition of satellite rejection and accumulation to obtain the pseudo-GNSS signal. The signal is used to compensate for the observed value in the integrated system. The model takes the advantages of LSTM, which is good at processing time series, and uses PSO to obtain the optimal value of important hyperparameters efficiently. Meanwhile, the improved threshold function is used to denoise the IMU (inertial measurement unit) data, which improves the SNR (signal-to-noise ratio) of IMU outputs effectively. Finally, the performance of the algorithm is proved by actual road test. Compared with INS, the method can reduce the maximum errors of latitude and longitude by at least 98.78% and 99.10% while the satellite is lost for 60 s, effectively improving the accuracy of the GNSS/INS system in satellite rejection.
- 2 July 2023
- Published by MDPI AG in Electronics
- Vol. 12 (13) , 2905
Open AccessAn GNSS/INS Integrated Navigation Algorithm Based on PSO-LSTM in Satellite Rejection
When the satellite signal is lost or interfered with, the traditional GNSS (Global Navigation Satellite System)/INS (Inertial Navigation System) integrated navigation will degenerate into INS, which results in the decrease in navigation accuracy. To solve these problems, this paper mainly established the PSO (particle swarm optimization) -LSTM (Long Short-Term Memory) neural network model to predict the increment of GNSS position under the condition of satellite rejection and accumulation to obtain the pseudo-GNSS signal. The signal is used to compensate for the observed value in the integrated system. The model takes the advantages of LSTM, which is good at processing time series, and uses PSO to obtain the optimal value of important hyperparameters efficiently. Meanwhile, the improved threshold function is used to denoise the IMU (inertial measurement unit) data, which improves the SNR (signal-to-noise ratio) of IMU outputs effectively. Finally, the performance of the algorithm is proved by actual road test. Compared with INS, the method can reduce the maximum errors of latitude and longitude by at least 98.78% and 99.10% while the satellite is lost for 60 s, effectively improving the accuracy of the GNSS/INS system in satellite rejection.