[ODE] How to realize natural interaction?

Jon Watte (ODE) hplus-ode at mindcontrol.org
Sun Apr 29 12:58:52 MST 2007


I think it would be better to realize that using mass on the bodies, and 
limiting the force you apply to the player body (and thus through the 
fixed joint). Yes, this means that the cursor must be detached from the 
mouse movement -- when holding a heavy body, the cursor will move a lot 
slower than when holding a light body (nor not holding anything at all).

Cheers,

          / h+


Roland Landertshamer wrote:
> Hi all,
>
> This is my first post to this list and since it probably will be not the last 
> one i think i should introduce myself. My name is Roland Landertshamer, i'm a 
> student and i'm working on my master thesis about physics simulation in 
> networked virtual environments.
> For the implementation of my work i'd like to use ODE as physics engine. In my 
> application world entities are represented by rigid bodies. The users of the 
> application should be able to interact with this objects in the sense of 
> grabbing an entity via a virtual hand and moving it around. The virtual hand 
> should only be visible and not collide with other rigid bodies. In my current 
> implementation i realized the virtual hand as a body without geometry to 
> ignore collisions. Before each simulation step the transformation of the 
> virtual hand is set manually and the velocities and forces are zeroed. When 
> the virtual hand moves into the bounding box of an entity the user can grab 
> this object. I realized the grabbing by adding a fixed joint between the 
> virtual hand and the entity. This works quite well so far. The simulation 
> fixes the joint error produced by the manual positioning of the virtual hand 
> quite well.
> My problem now is that i can interact with all objects independent of their 
> mass. For example i can grab an object with mass 1000 and throw it away the 
> same way as it would be with an object with mass 1. The desired behaviour 
> would be that an object with more mass should be harder to move and rotate. I 
> thougt about realizing this via a linear and angular spring. Any ideas how to 
> implement this in ODE?
>
> Roland
> _______________________________________________
> ODE mailing list
> ODE at ode.org
> http://ode.org/mailman/listinfo/ode
>
>
>   


More information about the ODE mailing list