#1
|
|||
|
|||
physics engine exception
I'm working on a robotics simulation with a robotic arm and different obstacles within a room. Collision detection is enabled for the obstacles and the arm, with the arm segments having viz.collide_notify enabled.
The program is set up such that there are multiple tasks, and in between each task, the obstacle objects are removed and new copies are added, since not all obstacles are used on every task. I keep encountering a physics engine exception between tasks, usually when the task ends with a collision taking place (successfully touching a target object with the end of the robotic arm is the current means for ending a task). The exception warning generally does not print out, however, until after all the old objects have been detroyed, the new objects have been created, and the main loop of the program has been reset. ** ERROR: Caught exception in physics engine I have tried disabling collision detection before the objects are destroyed and re-enabling it after the new ones are created with no success. I'm not sure if the program is actually this slow, or if it's just the print out capability, but when this problem occurs, the output indicates that the next task began before the program realized that the collision between an already-destroyed object and the arm ended. Initially, the exception warning only prints once, and but after a couple more tasks are completed, it continues to print each time the main loop runs, which makes continuing the program impossible. I was wondering if anyone had ever encountered this type of problem before and if there was a way to fix it. Thank you |
#2
|
|||
|
|||
The physics engine usually throws an exception when the number of contacts during a time-step exceed a certain amount. This typically occurs when you place a large number of physics objects at the same location. However, it could be related to something else. Do you have any small sample scripts that recreate this issue? If you prefer, you can email any sample directly to me at lashkari@worldviz.com
|
#3
|
|||
|
|||
Is there any way to find out exactly why a physics exception is thrown ?
I have a similar problems with Bowling pins which have COLLIDE_NOTIFY turned on resulting in ** ERROR: Caught exception in physics engine |
#4
|
|||
|
|||
Can you post an example showing this problem or email directly to loomis@worldviz.com?
|
Thread Tools | |
Display Modes | Rate This Thread |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Physics demos | mjabon | Vizard | 1 | 09-05-2007 01:00 PM |
physics documentation | vadrian | Vizard | 7 | 01-09-2006 10:16 AM |
another physics question | adimov | Vizard | 1 | 07-22-2004 03:56 PM |
Physics engine question | adimov | Vizard | 2 | 07-22-2004 12:33 PM |