Concurrent Path Planning for Humanoid Robot Joint Subsystems
Solution Overview
Problem
The control of multiple robotic joints in shared workspaces becomes complex due to interference zones, especially as the number of degrees of freedom increases, making existing serial handshaking protocols ineffective for coordinating motion in high DOF robotic systems.
Innovation Solution
A robotic system and control method that uses an algorithm to automatically group joints into task-specific subsystems, assign subtasks, and coordinate their execution asynchronously, employing a runtime engine with an asynchronous execution management module and a scheduling module to manage independent and synchronized tasks across multiple robots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If serial handshaking protocol is used to coordinate multiple robots, then communication between robots can be established, but the protocol loses effectiveness as the number of degrees of freedom increases
Solution Approach 1:
The controller segments the high-DOF robotic system into multiple independent lower-DOF virtual robots, each capable of autonomous path planning. This segmentation reduces the coordination complexity by allowing each segment to operate independently while maintaining overall task coordination, thus resolving the effectiveness loss in serial handshaking as DOF increases.
Solution Approach 2:
The patent introduces a new dimension of control by implementing concurrent path planning in the task space rather than relying on serial communication protocols in the control space. This dimensional shift allows multiple high-DOF robots to plan and execute paths simultaneously without the coordination overhead that plagues serial handshaking protocols.
2Productivity
If multiple robots operate in shared workspace, then cooperative work tasks can be executed, but interference zones complicate the control problem
Solution Approach 1:
The system performs preliminary path planning for all robots before execution, computing interference-free trajectories in advance. By planning paths concurrently and identifying potential interference zones beforehand, the system avoids real-time control complexity while maintaining productive cooperative operation in shared workspaces.
3Ease of operation
If conventional 6 DOF robot control is used, then control is considered routine, but high DOF robots require asynchronous and coordinated joint motion
Solution Approach 1:
The patent introduces an intermediary layer (the controller with concurrent path planning algorithm) that translates high-level task requirements into coordinated joint motions for high-DOF robots. This intermediary maintains ease of operation by abstracting away the complexity of asynchronous coordination, while still achieving the required human-like dexterity through precise multi-joint control.
Data Source
AI summary
A robotic system includes a controller and one or more robots each having a plurality of robotic joints. Each of the robotic joints is independently controllable to thereby execute a cooperative work task having at least one task execution fork, leading to multiple independent subtasks. The controller coordinates motion of the robot(s) during execution of the cooperative work task. The controller groups the robotic joints into task-specific robotic subsystems, and synchronizes motion of different subsystems during execution of the various subtasks of the cooperative work task. A method for executing the cooperative work task using the robotic system includes automatically grouping the robotic joints into task-specific subsystems, and assigning subtasks of the cooperative work task to the subsystems upon reaching a task execution fork. The method further includes coordinating execution of the subtasks after reaching the task execution fork.