Unit of Velocity from a Random Walk measured by an accelerometer

Click For Summary
SUMMARY

The discussion centers on the integration of a velocity random walk into a Kalman filter measurement equation, specifically the term "-g + v", where g is acceleration in m/s² and v is velocity random walk expressed in m/s/sqrt(hr). The primary issue identified is the incompatibility of units between acceleration and the given velocity random walk, indicating a potential error in the formulation of the equation. The suggestion is made to reconsider the placement of the random walk process, advocating for its inclusion in the state propagation equation instead of the measurement equation.

PREREQUISITES
  • Understanding of Kalman filter principles
  • Familiarity with units of measurement in physics, particularly acceleration and velocity
  • Knowledge of random walk processes in stochastic modeling
  • Experience with mathematical derivations involving unit conversions
NEXT STEPS
  • Research unit conversion techniques for velocity and acceleration
  • Learn about the implementation of random walk processes in state propagation equations
  • Study the mathematical foundations of Kalman filters and their applications
  • Explore common pitfalls in unit consistency within measurement equations
USEFUL FOR

Engineers, data scientists, and researchers working with Kalman filters, particularly those involved in sensor fusion and motion estimation tasks.

hoddy
Messages
4
Reaction score
0
TL;DR
Hi, I am working on a kalman filter where my measurement equation involves "-g + v" , where g is in m/s^2 and v is velocity random walk given in m/s/sqrt(hr). Feels like a stupid question, but how can I transform the unit of velocity random walk so I can do the calculation correctly?
Hi, I am working on a kalman filter where my measurement equation involves "-g + v" , where g is in m/s^2 and v is velocity random walk given in m/s/sqrt(hr). Feels like a stupid question, but how can I transform the unit of velocity random walk so I can do the calculation correctly?
 
Engineering news on Phys.org
hoddy said:
Summary: Hi, I am working on a kalman filter where my measurement equation involves "-g + v" , where g is in m/s^2 and v is velocity random walk given in m/s/sqrt(hr). Feels like a stupid question, but how can I transform the unit of velocity random walk so I can do the calculation correctly?

Hi, I am working on a kalman filter where my measurement equation involves "-g + v" , where g is in m/s^2 and v is velocity random walk given in m/s/sqrt(hr). Feels like a stupid question, but how can I transform the unit of velocity random walk so I can do the calculation correctly?
It looks to me as though you are trying to add quantities with incommensurate units. ##g## has units of acceleration, and ##v## has units that are neither velocity or acceleration. That suggests that there is an error somewhere in the derivation of the ##-g+v## term. Maybe you have assumed that some variable has a value of ##1##, and forgotten to carry along its units.
 
  • Like
Likes   Reactions: scottdave
Also, I wonder why you are including your random walk process in the measurement equation rather than in the state propagation equation?
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 2 ·
Replies
2
Views
4K
  • · Replies 3 ·
Replies
3
Views
729
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 47 ·
2
Replies
47
Views
5K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 5 ·
Replies
5
Views
4K
  • · Replies 4 ·
Replies
4
Views
6K
  • · Replies 6 ·
Replies
6
Views
4K
Replies
2
Views
2K