Trained human-intention classifier for safe and efficient robot navigation
US-9776323-B2 · Oct 3, 2017 · US
US10912253B2 · US · B2
| Field | Value |
|---|---|
| Publication number | US-10912253-B2 |
| Application number | US-201715711467-A |
| Country | US |
| Kind code | B2 |
| Filing date | Sep 21, 2017 |
| Priority date | Sep 22, 2016 |
| Publication date | Feb 9, 2021 |
| Grant date | Feb 9, 2021 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
The invention regards a robotic gardening device comprising driving means for propelling the robotic gardening device, a working tool for performing dedicated gardening work and a controlling unit for controlling said driving means and the working tool and a method for controlling the same. The robotic gardening device further comprises at least one environment sensor generating a signal indicative of objects in the environment of the robotic gardening device, a computing unit for classifying these objects, wherein the classes comprise at least two different classes for objects being determined to be humans. The computing unit is configured to control the driving means and/or the working device according to a predetermined behavior associated with the respective objects class.
Opening claim text (preview).
The invention claimed is: 1. Robotic gardening device comprising driving means for propelling the robotic gardening device, a working tool for performing dedicated gardening work, a controlling unit for controlling said driving means and the working tool, at least one environment sensor generating a signal indicative of objects in the environment of the robotic gardening device, and a computing unit for classifying the objects in plurality of classes, wherein the classes comprise at least two different classes for objects being determined to be different kinds of humans that require a different treatment with respect to safety, wherein the computing unit is configured to predict a future behavior for a detected and classified human, and wherein the controlling unit is configured to control the driving means to avoid a collision between a human and the robotic gardening device or to move the robotic gardening device towards the human depending on the predicted behavior associated with the respective object class. 2. Robotic gardening device according to claim 1 , wherein the object classes comprise at least “adult” and “children”. 3. Robotic gardening device according to claim 2 , wherein the object classes further comprise “animal” or “unknown person”. 4. Robotic gardening device according to claim 1 , wherein the computing unit is configured to predict at least one of the following possible future behaviors: crossing the current path of the robotic gardening device, interact with the robotic gardening device, stop the robotic gardening device, steal the robotic gardening device, touch the robotic gardening device, break into a house, play in the garden, do garden work, mire the garden, damage the garden. 5. Robotic gardening device according to claim 1 , wherein the computing unit comprises a classifier using deep learning or HOG features or optical flow or stereo camera depth data for determining an objects as a human or an animal. 6. Robotic gardening device according to claim 1 , wherein garden map data is stored in a memory of the robotic gardening device and the computing unit is configured to read in the map data and use it for classifying objects. 7. Robotic gardening device according to claim 1 , wherein predetermined behaviors include at least one or a plurality of: turning away from an obstacle, driving around an obstacle, returning to a base station, driving to a different zone, driving speed change, turn off working tool, follow the obstacle, approaching the obstacle, record sensor data, performing a scaring action, pausing and warning, reschedule working operation. 8. Robotic gardening device according to claim 1 , wherein a plurality of predetermined behaviors is associated with one objects class and the computing unit is configured to select one out of the plurality of predetermined behaviors depending on object distance, daytime, number of life form objects detected at the same time, position of a life form object. 9. Robotic gardening device according to claim 1 , wherein the robotic gardening device comprises an interface for receiving user input defining association of predetermined behaviors and object classes. 10. Robotic gardening device according to claim 1 , wherein the robotic gardening device comprises an interface for outputting an information signal indicative of presence of a life form object in a configurable time period. 11. Robotic gardening device according to claim 1 , wherein the robotic gardening device comprises indication means to indicate its own future behavior. 12. Robotic gardening device according to claim 1 , wherein the robotic gardening device is an autonomous lawn mower or an autonomous scarifier. 13. Method for controlling a robotic gardening device comprising the steps of: sensing the environment around the robotic gardening device, determining if perceived objects are human and if yes, classifying detected humans that require a different treatment with respect to safety in one of at least two different classes, predicting a future behavior for a detected and classified human, determining a behavior to be performed by the robotic gardening device according to the predicted future behavior associated with the respective object class, and controlling a driving means to avoid a collision between a human and the robotic gardening device or to move the robotic gardening device towards the human depending on the predicted behavior associated with the respective object class. 14. Robotic gardening device according to claim 1 , wherein the computing unit is configured to control the working tool according to the predicted behavior associated with the respective object class.
Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands · CPC title
Terrestrial scenes (scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59) · CPC title
Avoiding collisions with obstacles · CPC title
Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units · CPC title
involving models · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.