Depth-Sensor Servo Control for Unknown Object Grasping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing operating systems that employ feedback control via visual servos face challenges in accurately positioning and orienting robotic hands to grasp and manipulate objects, especially when feature extraction is difficult or when objects are unknown, as they require precise image processing and feature detection.
Innovation Solution
An operating system that includes a target surface detection unit, position calculation unit, and movement control unit, utilizing depth images from a depth sensor to detect target surfaces, calculate precise positions and directions, and control actuators to minimize positional and directional deviations, allowing for reliable object grasping and manipulation without the need for explicit image features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If visual servo feedback control is used with traditional image processing, then the system can control robotic movement, but it fails to accurately position and orient robotic hands when feature extraction is difficult
Solution Approach 1:
The patent replaces traditional visual servo feedback control that relies on image feature extraction with a depth sensor-based control system. The depth sensor directly provides three-dimensional position information, eliminating the need for complex image processing and feature detection, thereby achieving accurate positioning without being constrained by feature extraction difficulties
Solution Approach 2:
The patent changes the measurement parameter from two-dimensional image features to three-dimensional depth information. By using depth sensors to directly measure distance and position data, the system transforms the control input from visual features to spatial coordinates, enabling accurate positioning and orientation without relying on image processing
2Productivity
If traditional visual servo control is used, then the system can execute feedback control, but it requires precise image processing and feature detection which is not feasible for unknown objects
Solution Approach 1:
The patent substitutes the complex image processing and feature detection system with a depth sensor-based measurement system. The depth sensor directly provides spatial information needed for feedback control, eliminating the need for complex visual processing while maintaining control execution capability
Solution Approach 2:
The patent extracts only the essential spatial information (depth, position, orientation) from the environment using depth sensors, removing the unnecessary complexity of full image processing and feature detection. This extraction approach provides sufficient data for control execution without the burden of processing complete visual information
3Adaptability or versatility
If depth image data is used for control, then the system can grasp unknown objects without feature extraction, but additional depth sensing hardware is required
Solution Approach 1:
The patent makes the depth sensor serve multiple functions: it provides both positioning information and orientation information for feedback control, and enables handling of unknown objects without requiring separate feature detection systems. This multi-functionality increases adaptability while managing hardware complexity
Data Source
AI summary
An operating system according to an embodiment includes a target surface detection unit, a position calculation unit, a direction calculation unit, and a movement control unit. The target surface detection unit detects a target surface of a target object from a depth image obtained by a depth sensor. The position calculation unit calculates a first position for the detected target surface. The direction calculation unit calculates a first direction for the detected target surface. The movement control unit controls an actuator so as to reduce a positional deviation between a second position fixed with respect to the movable member and the first position and to reduce a directional deviation between a second direction fixed with respect to the movable member and the first direction.