-
Notifications
You must be signed in to change notification settings - Fork 43
Error Description
In the folder koptplanner/data/ a file “report.log” will be generated at runtime. The file lists a summary of the scenario and any error that occurs during execution. An explanation of the errors that can occur is given below:
Too many obstacles in the sampling area (OBSTACLE_INFEASIBILITY): The problem could be related with the clustering of obstacles in a certain place. Consider increasing max_obstacle_depth. The corner coordinates of the specific triangle, where the error occurred are given. Remark: It is normal, that obstacles dramatically reduce the efficiency of path computation.
No feasible position to inspect triangle (VIEWPOINT_INFEASIBILITY): The given triangle cannot be inspected with the current set of parameters. This might mean that either the triangle is too big or the camera constraints are too constraining. Remedy: Increase the minimum distance, use a wider field of view or adjust the mesh resolution.
No feasible heading to inspect triangle (VIEWPOINT_INFEASIBILITY): The given triangle cannot be inspected with the current set of parameters. This usually means, that the minimum distance is too small to allow inspection of the whole triangle. Remedy: Increase the minimum distance or adjust the mesh resolution.
No feasible connection has been found (CONNECTION_INFEASIBILITY): No valid states can be sampled by the RRT* planner.
A necessary parameter is not specified (MISSING_PARAMETER): One or more required parameters are not loaded by the roslaunch file.
No start position has been specified (NO_START_POSITION): At least one required pose has to be specified, corresponding to the the start position for the inspection.
Too few inspection areas have been specified (TOO_FEW_INSPECTION_AREAS): The minimum triangles in the mesh are 4.
The shape of an obstacle is invalid (INVALID_OBSTACLE_SHAPE): Only cuboidal obstacles are accepted (shape_msgs::SolidPrimitive::BOX).
Contact details: Andreas Bircher, Kostas Alexis, Autonomous Systems Lab