ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question

Revision history [back]

I'm sure the math could be derived to handle that better, but if you're going to use a generic filter package like this one, you're probably better off "modeling" kinematic constraints in the sensor data, as you did. If your robot can't move laterally, zero out the variables in question and feed them to the filter. In any case, I definitely don't have the cycles to add that feature, but if you want to take a crack at it and can do it in a way that doesn't break or fundamentally alter the filter's behaviour, I'd be willing to accept a PR.