Robot Motion Planning with Offline Plans and Online Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Motion planning in robotics is computationally expensive, leading to lag in reacting to real-time changes in the environment, and existing methods struggle to ensure compatibility with other moving robots, potentially causing collisions.
Innovation Solution
A system that pre-generates candidate motion plans offline and adapts them online based on real-time observations, using a library of motion plans to select and modify the most suitable plan to ensure compatibility and avoid collisions, while maintaining computational efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If motion planning is computed from scratch based on present observations, then the motion plan can be adapted to real-time changes, but the computation time is too long causing lag
Solution Approach 1:
The system pre-generates multiple candidate motion plans offline before real-time execution. During online operation, the robot selects from these pre-computed plans based on current observations, avoiding the need to compute a motion plan from scratch in real-time. This preliminary action resolves the contradiction by shifting computation to offline time while maintaining real-time adaptability through selection and adaptation of pre-generated plans.
Solution Approach 2:
The motion planning process is segmented into offline generation of multiple candidate plans and online selection/adaptation of the most suitable plan. This segmentation allows computationally intensive planning to be done offline, while online operations focus on lighter-weight selection and adaptation tasks, thus reducing real-time computation time while preserving adaptability.
2Productivity
If motion plan is computed without verifying compatibility constraints, then computation is faster, but collisions with other robots may occur
Solution Approach 1:
Compatibility constraints such as collision avoidance are verified in advance during the offline generation of candidate motion plans. The pre-generated plans are screened to ensure they satisfy compatibility constraints with other robots and environmental obstacles. During online execution, the robot selects from these pre-verified plans, maintaining both high computation speed and reliability for collision avoidance.
Solution Approach 2:
By pre-verifying compatibility constraints during offline plan generation, the system can skip the time-consuming compatibility verification step during online real-time operation. This allows the robot to rapidly select and execute motion plans without performing exhaustive collision checks in real-time, thus maintaining both speed and safety.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing online robotic motion planning from pre-generated motion plans. A library of pre-generated motion plans for performing a particular task is maintained. Each pre-generated motion plan comprises a plurality of waypoints and one or more actions. One or more present observations of a robot in a workcell are obtained. The one or more observations are classified. A pre-generated candidate motion plan that matches the labels assigned to the present observations of the robot in the workcell is selected from the library of pre-generated motion plans. The pre-generated candidate motion plan is adapted according to the present observations of the robot in the workcell to generate a final motion plan to be executed by the robot.