Robot Motion Planning With Staging Poses in Shared Workcells
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current motion planning techniques for robots in shared workspaces face challenges such as low work throughput due to idle robots, high computational time for collision avoidance, and inefficiencies in handling changing environments, especially when operating multiple high-degree-of-freedom robots.
Innovation Solution
The implementation of staging poses and dynamic motion planning algorithms that allow robots to autonomously adjust their positions and paths in real-time, using kinematic models and communication between robots to avoid collisions and optimize task completion without the need for extensive programming or perception sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If task-level de-confliction is used to operate multiple robots in a shared workspace, then collision-free operation is achieved, but work throughput decreases due to idle robots
Solution Approach 1:
The system dynamically adjusts robot motion plans in real-time based on the actual positions and motions of other robots. Instead of static task-level de-confliction where robots wait for flags to change, the motion planner continuously evaluates collision risk and generates updated trajectories, allowing robots to maintain motion and productivity while ensuring safety through real-time collision assessment.
2Productivity
If offline planning in combined joint space is used to achieve higher work throughput, then productivity increases, but computational time becomes extremely long for 12 DOF problems
Solution Approach 1:
The system segments the complex multi-robot motion planning problem into individual robot planning problems. Each robot's motion is planned independently in its own joint space rather than solving the combined joint space problem. The motion planner for each robot considers the motion plans of other robots as dynamic constraints, dividing the 12 DOF problem into manageable 6 DOF sub-problems that can be solved efficiently in real-time.
Solution Approach 2:
The system performs preliminary action by having each robot submit its desired motion plan in advance, allowing the motion planner to evaluate collision risk before execution. This enables real-time collision detection and trajectory adjustment without requiring exhaustive offline planning, as the system proactively checks and resolves potential conflicts before they occur during runtime.
3Reliability
If iterative simulation is used to optimize motion for multiple robots, then collision-free operation is ensured, but computation time increases to many hours
Solution Approach 1:
The system replaces iterative mechanical simulation with direct computational evaluation of collision risk. Instead of simulating robot motions step-by-step through physical interaction models, the motion planner uses analytical collision detection algorithms that directly compute whether trajectories intersect, substituting time-consuming iterative simulation with efficient mathematical evaluation that runs in real-time.
4Reliability
If manual optimization of robot motion is performed, then collision avoidance is achieved, but programming effort and complexity increase significantly
Solution Approach 1:
The system implements self-service by enabling robots to autonomously generate and adjust their own motion plans without requiring manual programming or optimization. The motion planner automatically evaluates collision risk and computes safe trajectories based on the robot's task goals and the real-time states of other robots, eliminating the need for engineers to manually optimize each robot's motion while maintaining collision-free operation.
Data Source
AI summary
The structures and algorithms described herein employ staging poses to facilitate the operation robots operating in a shared workspace or workcell, preventing or at least reducing the risk of collision while efficiently moving robots to one or more goals to perform respective tasks. Motion planning can be performed during runtime, and includes identifying one or more staging poses for a robot to advantageously position or configure a robot whose path is blocked or is expected to be blocked by one or more other robots, monitoring the other robots and moving the robot toward a goal in response to the path becoming unblocked or cleared. The staging pose can be identified using various heuristics to efficiently position or configure the robot to complete its task one its path becomes unblocked or cleared.