Robust Visual-Inertial Odometry Based on Deep Learning and Extended Kalman Filter

Zuo, Siqi; Shen, Kai; Zuo, Jianwen

doi:10.1109/cac53003.2021.9727349

Cited by 3 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, this study suggested unresolved challenges of motion estimation tasks, multi-sensor fusion under data absence, and data prediction in visual degradation scenarios. Subsequently, researchers aim to learn VO positions from raw image streams using CNN-LSTM [51,[53][54][55]. With the utilization of CNN-LSTM, some aimed to reduce IMU errors by predicting IMU dynamics in complex lighting conditions [54,55].…”

Section: Hybrid Fusion Enhanced By Aimentioning

confidence: 99%

See 1 more Smart Citation

Integrating GRU with a Kalman Filter to Enhance Visual Inertial Odometry Performance in Complex Environments

Tabassum,

Xu,

Petrunin

et al. 2023

Aerospace

View full text Add to dashboard Cite

To enhance system reliability and mitigate the vulnerabilities of the Global Navigation Satellite Systems (GNSS), it is common to fuse the Inertial Measurement Unit (IMU) and visual sensors with the GNSS receiver in the navigation system design, effectively enabling compensations with absolute positions and reducing data gaps. To address the shortcomings of a traditional Kalman Filter (KF), such as sensor errors, an imperfect non-linear system model, and KF estimation errors, a GRU-aided ESKF architecture is proposed to enhance the positioning performance. This study conducts Failure Mode and Effect Analysis (FMEA) to prioritize and identify the potential faults in the urban environment, facilitating the design of improved fault-tolerant system architecture. The identified primary fault events are data association errors and navigation environment errors during fault conditions of feature mismatch, especially in the presence of multiple failure modes. A hybrid federated navigation system architecture is employed using a Gated Recurrent Unit (GRU) to predict state increments for updating the state vector in the Error Estate Kalman Filter (ESKF) measurement step. The proposed algorithm’s performance is evaluated in a simulation environment in MATLAB under multiple visually degraded conditions. Comparative results provide evidence that the GRU-aided ESKF outperforms standard ESKF and state-of-the-art solutions like VINS-Mono, End-to-End VIO, and Self-Supervised VIO, exhibiting accuracy improvement in complex environments in terms of root mean square errors (RMSEs) and maximum errors.

show abstract

Section: Hybrid Fusion Enhanced By Aimentioning

confidence: 99%

“…Subsequently, researchers aim to learn VO positions from raw image streams using CNN-LSTM [51,[53][54][55]. With the utilization of CNN-LSTM, some aimed to reduce IMU errors by predicting IMU dynamics in complex lighting conditions [54,55].…”

Section: Hybrid Fusion Enhanced By Aimentioning

confidence: 99%

Integrating GRU with a Kalman Filter to Enhance Visual Inertial Odometry Performance in Complex Environments

Tabassum,

Xu,

Petrunin

et al. 2023

Aerospace

View full text Add to dashboard Cite

show abstract

“…However, both GPS and IMU have their limitations. GPS signals may be obstructed or interfered with, leading to inaccurate or intermittent position estimates [2] . IMU is prone to integration drift, resulting in an increasing error in attitude estimation over time.…”

Section: Introductionmentioning

confidence: 99%

Research on fusion algorithm based on multimode sensor

2024

Ninth International Symposium on Sensors, Mechatronics, and Automation System (ISSMAS 2023)

View full text Add to dashboard Cite

In vehicle position monitoring, the GPS system provides high accuracy but has a low refresh rate and is susceptible to instability in environments with weakened signals. The IMU can provide high-frequency motion data in a short time, but its integration errors accumulate over time, leading to position estimation drift. Combining the two through data fusion can compensate for their respective drawbacks and achieve better monitoring results. This paper explores GPS/IMU data fusion based on the principles of Kalman filtering, linear regression, and random forest. The results show that the Kalman filtering algorithm achieves better data fusion performance, with the Mean Squared Error (MSE) controlled below 0.68, while the MSE values obtained from linear regression and random forest algorithms are more than 6 times higher, indicating inferior performance. This confirms the superiority of the Kalman filtering approach over simple machine learning algorithms in data fusion. Additionally, the study simulates different noise environments and reveals that GPS signals have a significant impact on the fusion algorithm's results, being more susceptible to noise interference. This insight provides a feasible direction for further optimization of the Kalman filter.

show abstract

Integrating GRU with Kalman Filter to Enhance Visual Inertial Odometry Performance in Complex Environments

Tabassum,

Xu,

Petrunin

et al. 2023

Preprint

View full text Add to dashboard Cite

To enhance system reliability and mitigate vulnerabilities of the Global Navigation Satellite Systems (GNSS), it is common to fuse the Inertial Measurement Unit (IMU) and visual sensors with the GNSS receiver in the navigation system design, effectively enabling compensations with absolute position and reducing data gaps. To address shortcomings of traditional Kalman Filter (KF) such as sensor error, imperfect nonlinear system model, and KF estimation error, a GRU-aided ESKF architecture is proposed to enhance positioning performance. This study conducts Failure Mode and Effect Analysis (FMEA) to priories and identify the potential faults in the urban environment facilitating the design of improved fault-tolerant system architecture. The identified primary fault events are data association error and navigation environment error during fault conditions of feature mismatch, especially in the presence of multiple failure modes. A hybrid federated navigation system architecture is employed using a Gated Recurrent Unit (GRU) to predict state increments for updating the state vector in the Error Estate Kalman Filter (ESKF) measurement step. The proposed algorithm’s performance is evaluated in a simulation environment in MATLAB under multiple visually degraded conditions. Comparative results provide evidence that the GRU-aided ESKF outperforms standard ESKF and state-of-the-art solutions like VINS-Mono, End-to-End VIO, and Self-Supervised VIO, exhibiting higher efficiency in complex environments.

show abstract

Robust Visual-Inertial Odometry Based on Deep Learning and Extended Kalman Filter

Cited by 3 publications

References 11 publications

Integrating GRU with a Kalman Filter to Enhance Visual Inertial Odometry Performance in Complex Environments

Integrating GRU with a Kalman Filter to Enhance Visual Inertial Odometry Performance in Complex Environments

Research on fusion algorithm based on multimode sensor

Integrating GRU with Kalman Filter to Enhance Visual Inertial Odometry Performance in Complex Environments

Contact Info

Product

Resources

About