Verlet integration - first iteration

  • Context: Undergrad 
  • Thread starter Thread starter CactusPie
  • Start date Start date
  • Tags Tags
    Integration
Click For Summary
SUMMARY

The discussion focuses on implementing Verlet integration for simulating projectile motion with air drag. The initial challenge is determining the value for r(t-Δt) during the first iteration. It is established that a quadratic estimation can be used for the first step, specifically r(t+Δt) = r(t) + v(t)Δt + ½ a(t)Δt², which maintains a global error of order 2, consistent with Verlet integration. Subsequent steps can then utilize the Verlet method effectively.

PREREQUISITES
  • Understanding of Verlet integration and its application in numerical simulations
  • Familiarity with basic physics concepts such as projectile motion and air drag
  • Knowledge of numerical methods including Euler and Runge Kutta techniques
  • Proficiency in programming for implementing simulation algorithms
NEXT STEPS
  • Research the implementation of Verlet integration in Python using libraries like NumPy
  • Explore alternative numerical integration methods for comparison, such as the Leapfrog method
  • Study error analysis in numerical methods to understand the implications of different integration techniques
  • Investigate advanced topics in physics simulations, including adaptive time-stepping techniques
USEFUL FOR

Physics simulation developers, numerical analysts, and anyone interested in implementing and optimizing integration techniques for dynamic systems.

CactusPie
Messages
2
Reaction score
0
I'm trying to make a simple simulation of a projectile motion with air drag. I have already implemented it using Euler and fourth order Runge Kutta methods. I am however unsure about Verlet integration. The equation goes as follows:

[tex]r(t + Δt) = 2r(t) - r(t-Δt) + a(t)Δt^2[/tex]

I don't really know what value should I use during first iteration for

[tex]r(t-Δt)[/tex]

Could someone please help me?
 
Physics news on Phys.org
Verlet is one of many "multistep" integration techniques that need past data. As you noted, that past data doesn't exist on the first step. Some other technique, ideally one with comparable error characteristics, needs to be used to bootstrap a multistep integrator. For verlet integration, just use a quadratic estimation for that very first step,
[tex]r(t+\Delta t) = r(t) + v(t)\Delta t + \frac 1 2 a(t)\Delta t^2[/tex]
This has a global error of order 2, which is the same as verlet.

From the second step onward you can use verlet integration.
 
Thank you very much, that really helped!
 

Similar threads

  • · Replies 9 ·
Replies
9
Views
4K
  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 2 ·
Replies
2
Views
4K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 2 ·
Replies
2
Views
8K
Replies
5
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K