Robot Arm Object Tracking for Blind-Area Drop Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing object detection methods in object sorting processes, such as those using photoelectric sensors, vision-based detection, and motion models, suffer from issues like missing detections, low accuracy, and unreliable results due to visual blind areas and complex motion model dependencies.

Innovation Solution

A method and apparatus for target object detection that utilizes an image capturing device to track objects in real-time, determine their position and height relative to turnover boxes, and estimate landing points without relying on photoelectric sensors or motion models, thereby reducing visual blind areas and improving accuracy and reliability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If photoelectric sensors are used for target object detection, then detection speed can be improved, but detection accuracy deteriorates due to visual blind areas

Engineering Contradiction:
Improvedetection speedVSAvoiddetection accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent combines photoelectric sensor detection with vision-based detection methods to create a hybrid detection system. The photoelectric sensors provide fast detection speed while the vision system compensates for blind areas and improves accuracy, resolving the contradiction between speed and accuracy.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces an image capturing device as an intermediary to capture target object images, which then serve as additional detection data to complement photoelectric sensor outputs. This intermediary vision system fills the blind areas and improves overall detection accuracy without sacrificing the fast response of photoelectric sensors.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If vision-based detection methods are used, then detection accuracy can be improved, but detection reliability deteriorates due to complex motion model dependencies

Engineering Contradiction:
Improvedetection accuracyVSAvoiddetection reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent extracts and removes the complex motion model dependency from the detection system. By using a simplified detection approach that relies on direct image capture and photoelectric sensor data rather than complex motion modeling, the system maintains high accuracy while improving reliability through reduced computational complexity and fewer assumptions.

Inventive Principle:
Principle #2Taking out (Extraction)

3Reliability

If multiple detection methods are combined, then detection reliability can be improved, but device complexity increases

Engineering Contradiction:
Improvedetection reliabilityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements a multi-functional detection system where the image capturing device serves multiple purposes: it captures target object images for vision-based detection, provides visual feedback for sorting decisions, and compensates for photoelectric sensor blind areas. This universal approach improves reliability without proportionally increasing complexity by making each component serve multiple functions.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS12614305B2Target object detection method and apparatus, and electronic device, storage medium and program
Publication Date: 2026.04.28 BEIJING JINGDONG QIANSHITECHNOLOGY CO LTD
  • US12614305B2 patent drawing
  • US12614305B2 patent drawing
  • US12614305B2 patent drawing

AI summary

Provided in the embodiments of the present application are a target object detection method and apparatus, and an electronic device and a computer storage medium. The method comprises: when it is determined that a robot arm picks up a target object from any turnover box and the height of the target object is greater than the height of the turnover box, performing target tracking on the target object to obtain real-time position information of the target object; and when it is determined, according to the real-time position information of the target object, that the target object falls from the robot arm, determining an area in which the target object is located according to position information of the target object at the current moment and area position information, wherein the area position information comprises areas in which various types of turnover boxes are located.