Touch Panel Robot Teaching with Posture-Based Coordinate Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current control systems for robots lack effective manipulability during the teaching process, making it difficult for operators to intuitively input precise motion commands, especially with multi-articular arms that require complex positional and postural adjustments.
Innovation Solution
An operation device with a touch panel and posture sensor that outputs operation images, detects the operator's posture, and converts motion commands into a coordinate system suitable for the robot's multi-articular arm, allowing for intuitive point operations to specify motion directions and scalar quantities based on positional relationships on the screen.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a traditional teaching pendant is used to control robot motions, then the robot can be operated, but the manipulability and intuitiveness for teaching complex multi-articular arm motions is poor
Solution Approach 1:
The patent replaces traditional mechanical teaching pendant controls with a touch panel interface that detects operator posture through sensors. This substitution allows intuitive touch-based input where operators can simply touch screen locations to define motion directions and magnitudes, eliminating the need to manipulate complex physical controls while teaching complex multi-articular arm motions.
Solution Approach 2:
The patent introduces a coordinate system conversion mechanism as an intermediary between the operator's intuitive touch input and the robot's motion execution. The system automatically converts touch coordinates from the screen's coordinate system to the robot's workspace coordinate system, bridging the gap between simple operator actions and complex robot motions without requiring the operator to understand robotic coordinate systems.
2Measurement precision
If precise motion commands are input through traditional controls, then accurate robot positioning is achieved, but the operator cannot intuitively specify motion direction and magnitude simultaneously
Solution Approach 1:
The patent merges the specification of motion direction and magnitude into a single touch operation. When an operator touches a location on the screen, the system simultaneously determines both the direction (from the reference point to the touched point) and the magnitude (distance between these points) of the motion command, eliminating the need for separate controls for each parameter.
Solution Approach 2:
The patent adds a spatial dimension to the input interface by using the touch panel's two-dimensional screen space. The position where an operator touches the screen directly encodes both directional information (angle from reference) and magnitude information (distance from reference point), allowing precise two-parameter specification through a single spatial action rather than multiple sequential inputs.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances manipulability by allowing operators to intuitively set movement directions and scalar quantities, minimizing discrepancies between intended and actual inputs, and improving the teaching of complex robotic motions.
Implementation Method 1
a device posture detector configured to detect a posture of the operation inputter in a first coordinate system that is for controlling the robot
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An operation device includes an inputter for operating a robot. An outputter outputs an image to the inputter. A detector detects a posture of the inputter in a first coordinate system. Based on the posture, a turner turns a second coordinate system. Upon specification of a point on the image, a generator determines a motion direction of a leading end of the robot in the second system in correlation with a positional relationship between the one point and a reference point on the image; determines a motion scalar quantity of the leading end in correlation with a distance between the one point and the reference point; and generates a motion command including the motion direction and the motion scalar quantity. A convertor converts the command into a first-coordinate-system motion command. A command outputter outputs a control command.