METHOD FOR GENERATING A REPRESENTATION AND SYSTEM FOR TEACHING AN AUTONOMOUS DEVICE OPERATING BASED ON SUCH REPRESENTATION
A method for generating a representation based on which an autonomous device operates in its working environment at first generates a map of the working environment of the autonomous device by a mapping device. Then at least one area is defined in the map and at least one label is associated with each area to define the area's type by a map processing unit. A representation of the working environment is generated based on the at least one area and its associated label(s) by a representation generating unit and the representation is transmitted to the autonomous device.
1 . A method for generating a representation based on which an autonomous device operates in its working environment, the method comprising the following steps:
generating a map of the work environment of the autonomous device by a mapping device;
defining at least one area in the map and associating at least one label with each area to define the area's type, time information or a working parameter by a map processing unit;
generating a representation of the working environment based on the at least one area and its associated label(s) by a representation generating unit; and
transmitting the representation to the autonomous device.
2 . The method according to claim 1 , wherein
the at least one area is defined by an automated segmentation process.
3 . The method according to claim 1 , wherein
the map processing unit receives a human input to define or adapt an area.
4 . The method according to claim 1 , wherein
an input is received by the map processing unit indicating an area's type and the map processing unit associates a label corresponding to the type with the respective area.
5 . The method according to claim 1 , wherein
the labels correspond to at least the following types of areas: “collaborative work area”, “robot work area”, “human only work area”, “robot drive area”, “robot access inhibited area”, “water area”, “grass area”, “walkway area”, “paved area”, “front yard area”, “backyard area”, “farmyard area”, “leisure area”, time information or a working parameter.
6 . The method according to claim 1 ,
areas are divided into subareas, wherein different subareas of a common area are differently labelled.
7 . The method according to claim 1 , wherein
for generating the representation an intermediate representation is visualized to at least one human and feedback is received and used to adapt, delete or generate label(s).
8 . The method according to claim 1 , wherein
for generating the representation an intermediate representation is visualized to at least one human and feedback is received and used to adapt, delete or generate area(s) of the intermediate representation.
9 . The method according to claim 1 , wherein
an updated representation is generated based on a newly generated map while maintaining the areas labels or based on adapted labels while maintaining the areas.
10 . The method according to claim 1 , wherein
at least one label associated to at least one area is activated or deactivated when a predetermined event occurs.
11 . The method according to claim 1 , wherein
areas or subareas are represented based on polygon points, meshes, grids, voxels, triangular meshes, GPS positions map data, learned or trained model or neural networks.
12 . A system for teaching an autonomous device comprising:
an autonomous device and a representation generating system for generating a representation of a working environment of the autonomous device,
the representation generating system comprising
a mapping device for generating a map of the working environment,
a map processing unit for defining at least one area in the map and for associating at least one label with at least one of the areas to define an area's type, time information or a working parameter, and
a communication unit for transmitting the representation to the autonomous device.
13 . The system for teaching an autonomous device according to claim 12 , wherein
the map processing unit is configured to run an automated segmentation routine on the map.
14 . The system for teaching an autonomous device according to claim 12 , wherein
the representation generating system comprises an input/output device for outputting the map or an intermediate representation and for receiving a human input, wherein the input/output device is connected at least to the map processing unit.
15 . The system for teaching an autonomous device according to claim 12 , wherein
the map processing unit is configured to adapt, delete or generate areas based on received human feedback.
16 . The system for teaching an autonomous device according to claim 12 , wherein
the map processing unit is configured to adapt or delete associated labels or generate new labels based on received human feedback.
17 . The system for teaching an autonomous device according to claim 12 , wherein
the representation generating unit is configured to generate an intermediate representation and forward such intermediate representation to the input/output device.
18 . The system for teaching an autonomous device according to claim 12 , wherein
the map processing unit is configured to combine a new map with existing areas and their labels.
19 . The system for teaching an autonomous device according to claim 12 , wherein
the representation generating system comprises an augmented reality device, in particular an augmented reality wearable device, a smartphone or a tablet computer.
20 . The system for teaching an autonomous device according to claim 12 , wherein
the autonomous device is an autonomous lawnmower, a vacuum cleaner, an industrial robot, a welding robot, a drone or an autonomous vehicle.