Skip to content

Planning to a target in unknown space #11

Description

@MartinSegaert

Hello,

In docs/report/UnifiedAutonomyStack-Outline.md it is stated :

2.3.2 Collision-free Planning to a Target

The Unified Autonomy Stack facilitates planning to a desired waypoint both within the already explored space, as well as in the unknown, as long as this is iteratively found to be possible. For the first goal, the method simply exploits its local and global graphs to plan a safe path to any of the explored locations. For the latter, the method plans to a waypoint close to the frontier of the explored volume and in the direction of the desired target, while it then iteratively repeats this process as the robot unveils more of the space based on its onboard depth sensing.

When using the default "plan to waypoint" button in the planner interface, the planner is only able to reach positions inside the global graph.
Is planning to a target in unknown space possible with the current implementation of the Unified Autonomy Stack (more specifically its planner) ? Is it something that is in development or maybe already developed but not made open source yet ? Or is it available somewhere/somehow that I am not aware ?

I opened a similar issue on the github of the GBPlanner : ntnu-arl/gbplanner_ros#71

Thank you for your help,

Have a nice day,

Martin

Metadata

Metadata

Labels

No labels
No labels

Type

No type

Fields

No fields configured for issues without a type.

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions