Table 1 Overview of recently related work about trajectory prediction using LSTM
Tasks | Reference | Methodology | Results | Publication date |
|---|---|---|---|---|
Walking | Zaroug et al. [19] | Encoder–decoder LSTM | Correlation in the order of 0.98 between predicted and actual trajectory | 2020 |
Walking | Su et al. [20] | LSTM | A correlation of 0.98 in the predicted trajectory and 95% accuracy in phase prediction | 2020 |
Walking & running | Hernandez et al. [21] | DeepConvLSTM | MAE in range 2.2(0.9)–5.1(2.7) degrees | 2021 |
Walking | Jia et al. [22] | LSTM | RMSE in rage 0.348–0.713 degrees, correlation in the order of 0.99 | 2021 |
Walking | Zarough et al. [23] | LSTM | NRMSE in range 2.82–5.31% | 2021 |
Walking | Zhu et al. [24] | Attention-based CNN–LSTM | Within a predicted horizon of 60 ms, the prediction RMSE is as low as 0.317 degrees | 2021 |
Walking | Challa et al. [25] | LSTM | The gait trajectories obtained through the proposed model are also validated on the HOAP-2 robot simulator | 2022 |
Walking | Semwal et al. [26] | LSTM–CNN | A high correlation of 0.98 between the actual and the predicted trajectories, and an R-2 Score of 0.94 is obtained | 2023 |
Walking | Romero-Sorozábal, et al. [27] | LSTM & Regression | RMSE of 13.40 mm and a correlation coefficient of 0.92 for the regression model, and RMSE of 12.57 mm and a correlation of 0.99 for the LSTM | 2024 |