Robot Control Apparatus Singular Point Trajectory Generation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing robot control systems face challenges in accurately controlling robot movement near singular points, where infinite solutions exist for joint movement, leading to movement inhibiting regions, and struggle to synchronize image feature changes with robot control cycles due to low frame rates of general-purpose cameras.

Innovation Solution

A robot control apparatus that includes an actuator, joint shaft angle sensor, image capturing unit, image feature amount calculating and estimating units, image Jacobian calculating unit, target trajectory generating unit, and torque command value calculating unit, which temporally interpolates image feature amounts, calculates image Jacobians, and adjusts joint shaft angles to generate a stable target trajectory while considering hardware constraints, thereby avoiding singular points and ensuring stable visual feedback.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If a target value is set near a singular point to achieve accurate position and posture control, then manufacturing precision is improved, but the robot may enter a movement inhibiting region causing loss of stability

Engineering Contradiction:
Improveposition and posture control accuracyVSAvoidcontrol stability
Core Design Contradiction:
Manufacturing precisionVSReliability

Solution Approach 1:

The system performs preliminary detection of singular points and pre-calculates alternative target values before the robot reaches problematic regions. By anticipating singular point encounters and preparing corrected trajectories in advance, the system prevents entry into movement inhibiting regions while maintaining accurate position and posture control throughout the motion sequence.

Inventive Principle:
Principle #10Preliminary action

2Productivity

If the robot moves quickly along the target trajectory to improve productivity, then productivity is improved, but control stability deteriorates near singular points due to rapid joint shaft angle velocity changes

Engineering Contradiction:
Improvetrajectory traversal speedVSAvoidcontrol stability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system dynamically adjusts the robot's motion speed based on real-time proximity to singular points. When approaching detected singular points, the system automatically reduces joint shaft angle velocity to prevent entering movement inhibiting regions, while maintaining higher speeds in safe regions. This dynamic speed modulation ensures both high productivity during normal operation and stable control near critical regions.

Inventive Principle:
Principle #15Dynamics

3Measurement precision

If the frame rate of the camera is increased to improve measurement precision of image features, then measurement precision is improved, but device complexity and cost increase

Engineering Contradiction:
Improveimage feature capture rateVSAvoidcamera hardware requirements
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system introduces an image feature temporal interpolation unit that acts as an intermediary between the low frame rate camera and the high-speed robot control system. This unit calculates intermediate image feature values between captured frames using interpolation algorithms, effectively synthesizing high-rate measurement data from low-rate camera input. This approach achieves the measurement precision needed for accurate visual feedback without requiring expensive high frame rate camera hardware.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS8774967B2Robot control apparatus
Publication Date: 2014.07.08 KK TOSHIBA
  • US8774967B2 patent drawing
  • US8774967B2 patent drawing
  • US8774967B2 patent drawing

AI summary

According to an embodiment, a target trajectory that takes into account the hardware constraints of a robot is generated, based on results obtained by calculating, temporally interpolating, and estimating image feature amounts from a captured image.