Return
Improving odometric sensor performance by real-time error processing and variable covariance☆
DOI:10.1016/j.mechatronics.2023.103123.png)
Abstract
En 中文
This paper presents a new method to increase odometric sensor accuracy by systematic and non-systematic errors processing. Mobile robot localization is improved combining this technique with a filter that fuses the information from several sensors characterized by their covariance. The process focuses on calculating the odometric speed difference with respect to the filter to implement an error type detection module in real time. The correction of systematic errors consists in an online parameter adjustment using the previous information and conditioned by the filter accuracy. This data is also applied to design a variable odometric covariance which describes the sensor reliability and determines the influence of both errors on the robot localization. The method is implemented in a low-cost autonomous wheelchair with a LIDAR, IMU and encoders fused by an UKF algorithm. The experimental results prove that the estimated poses are closer to the real ones than using other well-known previous methods.
Keywords:
Sensor fusion
Odometry
Localization
Mobile robot
Pose estimation
Journal
IF:
3.1
Papers:
2.9K
Citations:
5.7K
Organization
Cited Papers
A Novel Online Approach for Drift Covariance Estimation of Odometries Used in Intelligent Vehicle Localization
SENSORS
IF3.5
Comparisons on Kalman-Filter-Based Dynamic State Estimation Algorithms of Power Systems
IEEE ACCESS
IF3.6
Structure and nucleotide sequence of the Bacillus subtilis phenylalanyl-tRNA synthetase genes
Biochimie
IF0
Augmented Kalman Filter Design in a Localization System Using Onboard Sensors With Intrinsic Delays
IEEE SENSORS JOURNAL
IF4.5
Optimization of the chiral separation of some 2‐arylpropionic acids on an avidin column by modeling a combined response
Chirality
IF0

