Neurintel (402) 438-2693 |    



Improved Velocity Measurement

Technology firm specializing in vehicle control.

In this system, the accuracy of distance travelled was of paramount importance. A primary method to measure distance traveled is to calculate the distance as a function of time and velocity. I.E. distance traveled equals velocity multiplied by time:

d = v * t

Using their unique skill sets of modeling and control theory, Neurintel was able to meet the firm's request of an improved velocity estimate to enhance their control systems.

Key Challenges
Ad hoc estimating and prediction schemes such as the commonly used Kalman filtering approach, which rely on tedious tuning and constant monitoring to be effective, were not allowed. It was requested that a modeling-based method should be used to improve the velocity measurement. This approach offered increased reliability and stability, though it also involved an in-depth development process and success was not guaranteed.

The proposed design took advantage of the combined use of vehicle models and a modified Luenberger observer, which is an advanced technique used in control theory. An observer models a real system in order to provide more accurate estimates of relevant states, such as velocities or headings, and can be integrated into an existing control system.

Results were gauged by the percentage that the true velocity of the vehicle differed from the measured velocity (the firm's current scheme) or from the estimated velocity (Neurintel's improved scheme). In simulations, Neurintel's improved estimate of the velocity had less than 2% of the error of the firm's current measurement. In addition, a scheme was developed to translate this observer design process to a broad array of similar types of vehicles.

P.O. Box 83342
Lincoln, NE 68501-3342
Office: (402) 438-2693
Home | Who We Are | What We Do | Who We Work With | Investigation | Contact Us
Copyright © 2018 Neurintel. All rights reserved | Website by Digital House