You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The proposed feature should implement a functionality that allows the information encoded in the ontology to be used for generating a simulation of an environment. I suggest a Gazebo simulation so that we can directly use the HSR there. For example, using the information in the C069 ontology, it should be possible to generate an environment in which the objects are placed so that they satisfy the constraints in the ontology.
Suggested solution
The generation component should solve a constraint satisfaction problem, where the constraints are defined by relations encoded in the ontology. Here, relations define feasible locations for the objects, such that intersections between feasible regions would define the rough location of an object in the environment.
Caveats
This is potentially a difficult problem since the ontology only encodes qualitative information about objects (e.g. one object is to the left of another). There are thus potentially many solutions to the generation problem, some of which don't fully correspond to the real environment. But that might also be fine, as it could be used to test a robot's reliability in slightly different environments.
The text was updated successfully, but these errors were encountered:
Feature description
The proposed feature should implement a functionality that allows the information encoded in the ontology to be used for generating a simulation of an environment. I suggest a Gazebo simulation so that we can directly use the HSR there. For example, using the information in the C069 ontology, it should be possible to generate an environment in which the objects are placed so that they satisfy the constraints in the ontology.
Suggested solution
The generation component should solve a constraint satisfaction problem, where the constraints are defined by relations encoded in the ontology. Here, relations define feasible locations for the objects, such that intersections between feasible regions would define the rough location of an object in the environment.
Caveats
This is potentially a difficult problem since the ontology only encodes qualitative information about objects (e.g. one object is to the left of another). There are thus potentially many solutions to the generation problem, some of which don't fully correspond to the real environment. But that might also be fine, as it could be used to test a robot's reliability in slightly different environments.
The text was updated successfully, but these errors were encountered: