Robot Motion Sequence Control Using Predefined Basic Motions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Machine learning-based control systems for robots face challenges in efficiently determining complex motion sequences and adjusting internal parameters, such as weights in neural networks, which hinders fine adjustments and increases processing time.
Innovation Solution
A control device that predefines basic motions and their parameters, allowing for the selection of optimal motion sequences and parameter adjustments, enabling faster processing and user-friendly adjustments by breaking down motion into predetermined primitives and sequences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If machine learning is used to determine motion sequence, then automation is improved, but processing time increases
Solution Approach 1:
The patent segments the motion determination process into two distinct parts: (1) a learning machine that determines the motion sequence from initial to target states, and (2) a parameter determination part that sets specific motion parameters for each basic motion in the sequence. This segmentation allows the learning machine to focus only on sequence selection rather than all motion details, reducing processing time while maintaining automation.
Solution Approach 2:
The patent performs preliminary action by pre-defining multiple basic motions and their associated motion parameters before the actual motion determination. The learning machine selects from these pre-prepared basic motions to construct the motion sequence, avoiding the need to learn all motion parameters from scratch, thus reducing processing time.
2Adaptability or versatility
If a large learned model is used to determine complex motion sequence, then adaptability is improved, but processing time increases
Solution Approach 1:
The patent divides the complex motion determination task into two segments: sequence determination by the learning machine and parameter determination by the parameter determination part. This segmentation reduces the computational burden on the learning machine, allowing it to handle complex motion sequences without excessive calculation time.
Solution Approach 2:
The patent extracts the parameter determination function from the learning machine and places it in a separate parameter determination part. This extraction allows the learning machine to focus solely on sequence selection, improving processing speed while the parameter determination part handles the detailed parameter settings for each basic motion.
3Extent of automation
If internal parameters of learned model are used, then automation is improved, but ease of operation deteriorates
Solution Approach 1:
The patent introduces motion parameters as an intermediary between the learned model's internal representations and the actual robot control. These motion parameters serve as a bridge that users can understand and adjust, translating the complex internal parameters of the learned model into meaningful control variables that affect basic motion execution.
Solution Approach 2:
The patent enables parameter changes by allowing users to directly adjust motion parameters for each basic motion in the sequence. This provides ease of operation by letting users modify specific motion characteristics without needing to understand or adjust the internal parameters of the learned model, while still maintaining automatic control through the learning machine's sequence determination.
Data Source
AI summary
This control device for controlling the motion of a robot comprises a first processing part and a command part. The first processing part sets a first state of the robot and a second state to which the robot transitions from the first state as inputs, and sets at least one basic motion selected from a plurality of basic motions the robot is instructed to perform for transitioning from the first state to the second state and the order in which the basic motions are to be performed as outputs. Prescribed operating parameters are set for each of the basic motions. The command part executes motion commands for the robot on the basis of the output from the first processing part.