Device and method for controlling a robotic device
OwnerMeng Guoinventor
Life of this patent
Estimating the term…
Abstract
A device and a method for controlling a robotic device, including a control model. The control model includes a robot trajectory model, which for the pickup includes a hidden semi-Markov model with one or multiple initial states, a precondition model, which for each initial state of the robot trajectory model includes a probability distribution of robot configurations before the pickup is carried out, and an object pickup model, which for a depth image outputs a plurality of pickup robot configurations having a respective associated probability of success.
Claims14
A method for controlling a robotic device, including a control model that is configured to control the robotic device for picking up an object, the control model including a robot trajectory model, a precondition model, and an object pickup model, the robot trajectory model for the pickup including a hidden semi-Markov model with one or multiple initial states and one or multiple end states, the precondition model including, for each initial state of the robot trajectory model, a probability distribution of robot configurations before the pickup is carried out, the object pickup model, in response to an input of a depth image, outputting a plurality of pickup robot configurations having a respective associated probability of success, the method comprising the following steps:
- ascertaining a depth image that shows one or multiple objects;
- for at least one object to be moved of the one or multiple objects, ascertaining one or multiple pickup robot configurations having a probability of success above a predefined threshold value with the aid of the object pickup model, using the depth image;
- ascertaining, for the pickup robot configuration of the one or multiple pickup robot configurations having a highest probability according to the precondition model, a robot trajectory for the pickup with the aid of the robot trajectory model;
- and
- controlling the robotic device for carrying out the pickup of the object using the ascertained robot trajectory.
A device configured to control a robotic device, including a control model that is configured to control the robotic device for picking up an object, the control model including a robot trajectory model, a precondition model, and an object pickup model, the robot trajectory model for the pickup including a hidden semi-Markov model with one or multiple initial states and one or multiple end states, the precondition model including, for each initial state of the robot trajectory model, a probability distribution of robot configurations before the pickup is carried out, the object pickup model, in response to an input of a depth image, outputting a plurality of pickup robot configurations having a respective associated probability of success, the device configured to:
- ascertain a depth image that shows one or multiple objects;
- for at least one object to be moved of the one or multiple objects, ascertain one or multiple pickup robot configurations having a probability of success above a predefined threshold value with the aid of the object pickup model, using the depth image;
- ascertain, for the pickup robot configuration of the one or multiple pickup robot configurations having a highest probability according to the precondition model, a robot trajectory for the pickup with the aid of the robot trajectory model;
- and
- control the robotic device for carrying out the pickup of the object using the ascertained robot trajectory.
A non-transitory computer-readable medium on which is stored instructions for controlling a robotic device, including a control model that is configured to control the robotic device for picking up an object, the control model including a robot trajectory model, a precondition model, and an object pickup model, the robot trajectory model for the pickup including a hidden semi-Markov model with one or multiple initial states and one or multiple end states, the precondition model including, for each initial state of the robot trajectory model, a probability distribution of robot configurations before the pickup is carried out, the object pickup model, in response to an input of a depth image, outputting a plurality of pickup robot configurations having a respective associated probability of success, the instructions, when executed by a processor, causing the processor to perform the following steps:
- ascertaining a depth image that shows one or multiple objects;
- for at least one object to be moved of the one or multiple objects, ascertaining one or multiple pickup robot configurations having a probability of success above a predefined threshold value with the aid of the object pickup model, using the depth image;
- ascertaining, for the pickup robot configuration of the one or multiple pickup robot configurations having a highest probability according to the precondition model, a robot trajectory for the pickup with the aid of the robot trajectory model;
- and
- controlling the robotic device for carrying out the pickup of the object using the ascertained robot trajectory.
Dependencies are inferred from the claim text.