Conversation
Is the cost when using an Ackermann motion model meant to be modified twice?
|
@ObeidHaidar, all pull requests must be targeted towards the |
|
Also @SteveMacenski , this probably isn't the right place to ask, but any recommendation for trying to increase the prediction horizon? The current horizon doesn't work as good when controlling robots with a bigger footpath. |
SteveMacenski
left a comment
There was a problem hiding this comment.
Please target the main branch as the bot indicates. I'll backport to Humble immediately but as a point of process to keep the git history clean it needs to go there first
| (std::move(out_of_max_bounds_motion) + | ||
| std::move(out_of_min_bounds_motion)) * | ||
| data.model_dt, {1}, immediate) * weight_, power_); | ||
| else { |
There was a problem hiding this comment.
Linting issue, if nothing else } else {
There was a problem hiding this comment.
i'll fix it and add pull request to main. Just wanted to double check that it's not done on purpose. I'll add a return instead of an else statement and will create a new pull request to main.
Sure, do it! You can increase the number of trajectory timesteps, though it comes obviously with proportionally higher computational costs. |
Is the cost when using an Ackermann motion model meant to be modified twice?
Basic Info
For Maintainers: