How can I track positions from acceleration values?

  • Context: Undergrad 
  • Thread starter Thread starter dokkdurs
  • Start date Start date
  • Tags Tags
    Acceleration Position
Click For Summary
SUMMARY

This discussion focuses on tracking positions from acceleration values using an accelerometer for a toy car project. The key formula provided for calculating position is x = x₀ + v₀t + (at²)/2, where x₀ is the initial position, v₀ is the initial velocity, a is the acceleration, and t is time. The conversation emphasizes the importance of numerical integration techniques, specifically the trapezoid rule, to convert acceleration data into velocity and position over time. Accuracy and sampling frequency of the accelerometer are critical factors affecting the results.

PREREQUISITES
  • Understanding of basic kinematics equations
  • Familiarity with numerical integration methods, particularly the trapezoid rule
  • Knowledge of accelerometer data interpretation
  • Basic programming skills for implementing calculations
NEXT STEPS
  • Research "Numerical integration techniques for motion tracking"
  • Explore "Accelerometer data filtering methods" to improve accuracy
  • Learn about "Sensor fusion techniques" for enhanced position tracking
  • Investigate "Real-time data processing in embedded systems" for practical applications
USEFUL FOR

This discussion is beneficial for hobbyists, robotics enthusiasts, and engineers working on motion tracking projects, particularly those utilizing accelerometers for position estimation.

dokkdurs
Messages
1
Reaction score
0
Hello

I'm new to this forum and i was wondering if anyone here can help me with a problem I'm having.

I am trying to plot a track on which a toy car is running.The voltage on the motor stay the same.An accelerometer feeds me values on x,y and z axis(the z values stay pretty much the same).My problem is this : tracking positions from acceleration values.I am receiving values in -5000 and +5000.I have tried some methods of plotting the track but with no real results.

Thank you and sorry if this threads violates any forum rules or if i am too noob
 
Physics news on Phys.org
Answer from also a n00b:

So your position is
[tex]x=x_0 + v_0 t + at^2/2[/tex]
where x0 is the initial position, v0 is the initial velocity (i guess you might want to set them to zeros), and then a is the acceleration that your device is giving, while t is the time. Does this help?

P.S. I've noticed you get x and y accelerations, but I would only expect x to be relevant, i.e. you only want to know how the toy car goes forward. Isin't this the case?
 
Last edited:
If the acceleration varies over time, you'll need to do some form of numerical integration to convert accelerations into velocities and positions. Accuracy and frequency of the accelerometer outputs will be an issue.

Given a set of accelerations versus time with time elapsed between sample of Δt, and initial velocity v[0], and posiion p[0], for each axis, then you can calculate velocities and positions based on average acceleration and velocity (trapezoid rule):

a[...], v[0], p[0] are given

v[1] = v[0] + 1/2 (a[0] + a[1]) Δt
p[1] = p[0] + 1/2 (v[0] + v[1]) Δt

v[2] = v[1] + 1/2 (a[1] + a[2]) Δt
p[2] = p[1] + 1/2 (v[1] + v[2]) Δt

...

This should get you close.
 
Last edited:

Similar threads

  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 13 ·
Replies
13
Views
3K
  • · Replies 21 ·
Replies
21
Views
2K
  • · Replies 10 ·
Replies
10
Views
2K
  • · Replies 29 ·
Replies
29
Views
4K
  • · Replies 8 ·
Replies
8
Views
6K
  • · Replies 10 ·
Replies
10
Views
2K
  • · Replies 12 ·
Replies
12
Views
19K
  • · Replies 1 ·
Replies
1
Views
2K