Robot Task Preparation Using Preplanned Constraints and Shared Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for controlling robots in shared control and supervised autonomy lack efficiency in planning and transitioning between modes, leading to prolonged planning times and potential task failures due to collisions or energy constraints.
Innovation Solution
A method that pre-plans object-dependent constraint conditions using integrated autonomy, allowing for fast and iterative feasibility testing, and switches seamlessly between shared control and supervised autonomy using a shared control module with user and automation modules.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the robot performs real-time planning for each task, then the robot can adapt to current environment, but the planning time is prolonged
Solution Approach 1:
The system performs preliminary planning in a preparation phase before the execution phase, pre-computing constraint conditions and feasible action sequences. This advance preparation reduces the planning time during actual task execution while ensuring reliability through pre-validated plans.
Solution Approach 2:
The planning process is divided into distinct phases: preparation phase for computing constraint conditions and execution phase for carrying out tasks. This segmentation allows complex planning to be done in advance while keeping execution simple and fast.
2Adaptability or versatility
If the robot uses shared control mode for user assistance, then the robot can benefit from user input, but the control complexity increases
Solution Approach 1:
A unified control module implements multiple control modes (shared control and supervised autonomy) using the same action representations and planning framework. This universal approach provides control mode flexibility while avoiding the complexity of separate control systems for each mode.
Solution Approach 2:
The patent uses homogeneous action representations across different control modes, allowing seamless transitions between shared control and supervised autonomy. The same planning and execution framework handles both modes, reducing control system complexity.
3Reliability
If the robot checks all possible actions for feasibility, then the robot avoids collisions and energy issues, but the planning time increases
Solution Approach 1:
The system performs preliminary feasibility checks in the preparation phase, computing constraint conditions that encode collision avoidance and energy constraints before execution. This advance checking ensures reliability while keeping execution fast by avoiding repeated feasibility checks during task performance.
Data Source
AI summary
The invention relates to a method for preparing and carrying out tasks by means of a robot, in which method, in a preparation phase at least one probable action goal is determined on the basis of surroundings information and taking into account a user command probability, at least one preparation action sequence is generated which is aimed at the at least one probable action goal, and the at least one preparation action sequence is carried out. The invention also relates to a robot for carrying out tasks, and to a computer program.