Analysis of State Estimation Drift on a MAV Using PX4 Autopilot and MEMS IMU During Dead-reckoning

Jang, Jong Tai; Santamaria-Navarro, Àngel; Lopez, Brett T.; Agha–mohammadi, Ali–akbar

doi:10.1109/aero47225.2020.9172736

Cited by 13 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These predictions diverge quickly from the actual state values because numerical integration is applied thrice, and the biases used for compensation are not perfectly accurate [31]. Biases even vary with time, so they need to be updated regularly.…”

Section: Px4's Extended Kalman Filter Algorithm (Ekf2)mentioning

confidence: 99%

GPS-Denied Navigation Using Low-Cost Inertial Sensors and Recurrent Neural Networks

AbdulMajuid,

Mohamady,

Draz

et al. 2021

Preprint

View full text Add to dashboard Cite

Autonomous missions of drones require continuous and reliable estimates for the drone's attitude, velocity, and position. Traditionally, these states are estimated by applying Extended Kalman Filter (EKF) to Accelerometer, Gyroscope, Barometer, Magnetometer, and GPS measurements. When the GPS signal is lost, position and velocity estimates deteriorate quickly, especially when using low-cost inertial sensors. This paper proposes an estimation method that uses a Recurrent Neural Network (RNN) to allow reliable estimation of a drone's position and velocity in the absence of GPS signal. The RNN is trained on a public dataset collected using Pixhawk. This low-cost commercial autopilot logs the raw sensor measurements (network inputs) and corresponding EKF estimates (ground truth outputs). The dataset is comprised of 548 different flight logs with flight durations ranging from 4 to 32 minutes. For training, 465 flights are used, totaling 45 hours. The remaining 83 flights totaling 8 hours are held out for validation. Error in a single flight is taken to be the maximum absolute difference in 3D position (MPE) between the RNN predictions (without GPS) and the ground truth (EKF with GPS). On the validation set, the median MPE is 35 meters. MPE values as low as 2.7 meters in a 5-minutes flight could be achieved using the proposed method. The MPE in 90% of the validation flights is bounded below 166 meters. The network was experimentally tested and worked in real-time.

show abstract

Section: Px4's Extended Kalman Filter Algorithm (Ekf2)mentioning

confidence: 99%

GPS-Denied Navigation Using Low-Cost Inertial Sensors and Recurrent Neural Networks

AbdulMajuid,

Mohamady,

Draz

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Then, the EKF filter can only be based on the fusion of data from the UAV’s own sensors. Depending on flight conditions and the number and type of sensors, the flight can continue or must be interrupted due to possible calculation errors of the EKF filter [ 2 , 19 , 33 , 34 , 35 ].…”

Section: Introductionmentioning

confidence: 99%

The Influence of Temporal Disturbances in EKF Calculations on the Achieved Parameters of Flight Control and Stabilization of UAVs

Szczepaniak,

Szlachetko,

Lower

2024

Sensors

View full text Add to dashboard Cite

This article investigates the causes of occasional flight instability observed in Unmanned Aerial Vehicles (UAVs). The issue manifests as unexpected oscillations that can lead to emergency landings. The analysis focuses on delays in the Extended Kalman Filter (EKF) algorithm used to estimate the drone’s attitude, position, and velocity. These delays disrupt the flight stabilization process. The research identifies two potential causes for the delays. First cause is magnetic field distrurbances created by UAV motors and external magnetic fields (e.g., power lines) that can interfere with magnetometer readings, leading to extended EKF calculations. Second cause is EKF fusion step implementation of the PX4-ECL library combining magnetometer data with other sensor measurements, which can become computionally expensive, especially when dealing with inconsistent magnetic field readings. This can significantly increase EKF processing time. The authors propose a solution of moving the magnetic field estimation calculations to a separate, lower-priority thread. This would prevent them from blocking the main EKF loop and causing delays. The implemented monitoring techniques allow for continuous observation of the real-time operating system’s behavior. Since addressing the identified issues, no significant problems have been encountered during flights. However, ongoing monitoring is crucial due to the infrequent and unpredictable nature of the disturbances.

show abstract

“…For these reasons, the potentially unreliable measurements from GPS, lidars or cameras are usually fused with the data from an Inertial Measurement Unit (IMU) [1]- [3], whose acceleration and angular velocity measurements can be used to perform dead reckoning 1 between the temporary failures of the position sensors/estimators. Measurements from consumer-grade IMUs, however, are typically corrupted by drifting biases, calibration errors and noise [4], causing large position drifts when integrated, and making IMU-only dead reckoning reliable for a short amount of time.…”

Section: Introductionmentioning

confidence: 99%

Airflow-Inertial Odometry for Resilient State Estimation on Multirotors

Tagliabue¹,

How²

2021

Preprint

View full text Add to dashboard Cite

We present a dead reckoning strategy for increased resilience to position estimation failures on multirotors, using only data from a low-cost IMU and novel, bio-inspired airflow sensors. The goal is challenging, since low-cost IMUs are subject to large noise and drift, while 3D airflow sensing is made difficult by the interference caused by the propellers and by the wind. Our approach relies on a deep-learning strategy to interpret the measurements of the bio-inspired sensors, a map of the wind speed to compensate for position-dependent wind, and a filter to fuse the information and generate a pose and velocity estimate. Our results show that the approach reduces the drift with respect to IMU-only dead reckoning by up to an order of magnitude over 30 seconds after a position sensor failure in non-windy environments, and it can compensate for the challenging effects of turbulent, and spatially varying wind.

show abstract

Analysis of State Estimation Drift on a MAV Using PX4 Autopilot and MEMS IMU During Dead-reckoning

Cited by 13 publications

References 9 publications

GPS-Denied Navigation Using Low-Cost Inertial Sensors and Recurrent Neural Networks

GPS-Denied Navigation Using Low-Cost Inertial Sensors and Recurrent Neural Networks

The Influence of Temporal Disturbances in EKF Calculations on the Achieved Parameters of Flight Control and Stabilization of UAVs

Airflow-Inertial Odometry for Resilient State Estimation on Multirotors

Contact Info

Product

Resources

About