Reliability Indexed Sensor Fusion and Its Application to Vehicle Velocity Estimation

2006 ◽  
Vol 128 (2) ◽  
pp. 236-243 ◽  
Author(s):  
Hyeongcheol Lee
Keyword(s):  
Sensor Fusion ◽  
Uniform Boundedness ◽  
Velocity Estimation ◽  
Test Results ◽  
Lateral Velocity ◽  
Angle Sensor ◽  
Multiple Sensors ◽  
Error Covariance ◽  
Vehicle Velocity ◽  
Rate Sensor

Reliability indexed sensor fusion (RISF) is a new estimation techique which uses process and measurement noise covariances as the reliability index in an adaptive Kalman filter framework. In RISF, noise covariances are assumed to be highly uncertain and determined by engineering knowledge. The uniform boundedness of the RISF with incorrect noise covariances is proved in the sense that the error covariance is bounded if specified conditions are satisfied. The RISF technique is then applied to the vehicle longitudinal and lateral velocity estimation. Multiple sensors, such as the whell speed sensors, the accelerometers, the yaw rate sensor, and the steering angle sensor, are used for the velocity estimation. Test results show the accuracy of the vehicle velocity estimation by the proposed RISF technique.

scholarly journals Vehicle Velocity Estimation Fusion with Kinematic Integral and Empirical Correction on Multi-Timescales

Energies ◽  
2019 ◽  
Vol 12 (7) ◽  
pp. 1242
Author(s):  
Jiangyi Lv ◽  
Hongwen He ◽  
Wei Liu ◽  
Yong Chen ◽  
Fengchun Sun
Keyword(s):  
Autonomous Vehicles ◽  
Estimation Error ◽  
Longitudinal Velocity ◽  
Estimation Method ◽  
Velocity Estimation ◽  
Model Parameters ◽  
Lateral Velocity ◽  
Empirical Correction ◽  
Vehicle Velocity ◽  
Wide Range

Accurate and reliable vehicle velocity estimation is greatly motivated by the increasing demands of high-precision motion control for autonomous vehicles and the decreasing cost of the required multi-axis IMU sensors. A practical estimation method for the longitudinal and lateral velocities of electric vehicles is proposed. Two reliable driving empirical judgements about the velocities are extracted from the signals of the ordinary onboard vehicle sensors, which correct the integral errors of the corresponding kinematic equations on a long timescale. Meanwhile, the additive biases of the measured accelerations are estimated recursively by comparing the integral of the measured accelerations with the difference of the estimated velocities between the adjacent strong empirical correction instants, which further compensates the kinematic integral error on short timescale. The algorithm is verified by both the CarSim-Simulink co-simulation and the controller-in-the-loop test under the CarMaker-RoadBox environment. The results show that the velocities can be accurately and reliably estimated under a wide range of driving conditions without prior knowledge of the tire-model and other unavailable signals or frequently changeable model parameters. The relative estimation error of the longitudinal velocity and the absolute estimation error of the lateral velocity are kept within 2% and 0.5 km/h, respectively.

scholarly journals Gas Detection and Identification Using Multimodal Artificial Intelligence Based Sensor Fusion

2021 ◽  
Vol 4 (1) ◽  
pp. 3
Author(s):  
Parag Narkhede ◽  
Rahee Walambe ◽  
Shruti Mandaokar ◽  
Pulkit Chandel ◽  
Ketan Kotecha ◽  
...  
Keyword(s):  
Sensor Fusion ◽  
Network Architecture ◽  
Cost Effective ◽  
Sensor Data ◽  
Dense Layer ◽  
Gaseous Emissions ◽  
Multiple Sensors ◽  
Single Output ◽  
Novel Approach ◽  
Single Sensor

With the rapid industrialization and technological advancements, innovative engineering technologies which are cost effective, faster and easier to implement are essential. One such area of concern is the rising number of accidents happening due to gas leaks at coal mines, chemical industries, home appliances etc. In this paper we propose a novel approach to detect and identify the gaseous emissions using the multimodal AI fusion techniques. Most of the gases and their fumes are colorless, odorless, and tasteless, thereby challenging our normal human senses. Sensing based on a single sensor may not be accurate, and sensor fusion is essential for robust and reliable detection in several real-world applications. We manually collected 6400 gas samples (1600 samples per class for four classes) using two specific sensors: the 7-semiconductor gas sensors array, and a thermal camera. The early fusion method of multimodal AI, is applied The network architecture consists of a feature extraction module for individual modality, which is then fused using a merged layer followed by a dense layer, which provides a single output for identifying the gas. We obtained the testing accuracy of 96% (for fused model) as opposed to individual model accuracies of 82% (based on Gas Sensor data using LSTM) and 93% (based on thermal images data using CNN model). Results demonstrate that the fusion of multiple sensors and modalities outperforms the outcome of a single sensor.

scholarly journals Sensor Fusion for Recognition of Activities of Daily Living

Sensors ◽  
2018 ◽  
Vol 18 (11) ◽  
pp. 4029 ◽  
Author(s):  
Jiaxuan Wu ◽  
Yunfei Feng ◽  
Peng Sun
Keyword(s):  
Time Series ◽  
Sensor Fusion ◽  
Daily Living ◽  
Single Point ◽  
Recognition System ◽  
Sensor Data ◽  
Multiple Sensors ◽  
Life Patterns ◽  
Functional Capabilities ◽  
Correction Technique

Activity of daily living (ADL) is a significant predictor of the independence and functional capabilities of an individual. Measurements of ADLs help to indicate one’s health status and capabilities of quality living. Recently, the most common ways to capture ADL data are far from automation, including a costly 24/7 observation by a designated caregiver, self-reporting by the user laboriously, or filling out a written ADL survey. Fortunately, ubiquitous sensors exist in our surroundings and on electronic devices in the Internet of Things (IoT) era. We proposed the ADL Recognition System that utilizes the sensor data from a single point of contact, such as smartphones, and conducts time-series sensor fusion processing. Raw data is collected from the ADL Recorder App constantly running on a user’s smartphone with multiple embedded sensors, including the microphone, Wi-Fi scan module, heading orientation of the device, light proximity, step detector, accelerometer, gyroscope, magnetometer, etc. Key technologies in this research cover audio processing, Wi-Fi indoor positioning, proximity sensing localization, and time-series sensor data fusion. By merging the information of multiple sensors, with a time-series error correction technique, the ADL Recognition System is able to accurately profile a person’s ADLs and discover his life patterns. This paper is particularly concerned with the care for the older adults who live independently.

A kinematic observer with adaptive dead-zone for vehicles lateral velocity estimation

2018 ◽  
Author(s):  
Luca De Pascali ◽  
Francesco Biral ◽  
Matteo Cocetti ◽  
Luca Zaccarian ◽  
Sophie Tarbouriech
Keyword(s):  
Dead Zone ◽  
Velocity Estimation ◽  
Lateral Velocity

Velocity estimation in laterally varying media

Geophysics ◽  
1982 ◽  
Vol 47 (6) ◽  
pp. 884-897 ◽  
Author(s):  
Walter S. Lynn ◽  
Jon F. Claerbout
Keyword(s):  
Taylor Series Expansion ◽  
Velocity Estimation ◽  
Lateral Resolution ◽  
Velocity Variation ◽  
Lateral Velocity ◽  
Depth Migration ◽  
Derivative Term ◽  
Fourth Derivative ◽  
Zero Offset ◽  
Lateral Variations

In areas of large lateral variations in velocity, stacking velocities computed on the basis of hyperbolic moveout can differ substantially from the actual root mean square (rms) velocities. This paper addresses the problem of obtaining rms or migration velocities from stacking velocities in such areas. The first‐order difference between the stacking and the vertical rms velocities due to lateral variations in velocity are shown to be related to the second lateral derivative of the rms slowness [Formula: see text]. Approximations leading to this relation are straight raypaths and that the vertical rms slowness to a given interface can be expressed as a second‐order Taylor series expansion in the midpoint direction. Under these approximations, the effect of the first lateral derivative of the slowness on the traveltime is negligible. The linearization of the equation relating the stacking and true velocities results in a set of equations whose inversion is unstable. Stability is achieved, however, by adding a nonphysical fourth derivative term which affects only the higher spatial wavenumbers, those beyond the lateral resolution of the lateral derivative method (LDM). Thus, given the stacking velocities and the zero‐offset traveltime to a given event as a function of midpoint, the LDM provides an estimate of the true vertical rms velocity to that event with a lateral resolution of about two mute zones or cable lengths. The LDM is applicable when lateral variations of velocity greater than 2 percent occur over the mute zone. At variations of 30 percent or greater, the internal assumptions of the LDM begin to break down. Synthetic models designed to test the LDM when the different assumptions are violated show that, in all cases, the results are not seriously affected. A test of the LDM on field data having a lateral velocity variation caused by sea floor topography gives a result which is supported by depth migration.

Sign in / Sign up

Export Citation Format

Share Document