Mobile Robot Moving Object Tracking with Dynamic Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for tracking a moving object in a congested environment with multiple objects, such as those using range sensors or image processing, face challenges in accuracy and reliability, especially when objects have irregular shapes or move at high speeds, and sensor fusion systems struggle to maintain performance beyond camera image processing.
Innovation Solution
A mobile robot system that combines image capturing, range sensing, and congestion degree measurement to dynamically adjust tracking parameters, using a fusion estimation result from image processing and range sensor outputs, along with congestion and reflection intensity data, to enhance tracking accuracy and reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensor fusion is used to combine multiple sensors for tracking, then tracking reliability is improved, but device complexity increases
Solution Approach 1:
The patent combines image processing results with laser range sensor data into a unified tracking system. The image processing unit identifies target objects and their types, while the range sensor provides distance measurements. These two independent sensing systems are merged through a sensor fusion mechanism that integrates their outputs to achieve more reliable tracking than either system could provide alone, directly resolving the contradiction between reliability improvement and complexity increase.
Solution Approach 2:
The tracking system is designed to handle multiple object types (vehicles, pedestrians, cyclists) and multiple sensing modalities (image processing and range sensing) through a single unified framework. The sensor fusion unit universally processes inputs from both image processing and range sensor regardless of object type, enabling the system to maintain reliability across diverse scenarios without requiring separate specialized systems for each case.
2Reliability
If image processing is used to specify target object type, then tracking accuracy is improved, but measurement precision deteriorates due to matching errors and resolution limits
Solution Approach 1:
The patent introduces laser range sensor data as an intermediary element that bridges the gap between image processing and precise position measurement. While image processing accurately identifies object types and specifications, the range sensor provides independent, high-precision distance measurements that are not affected by image processing errors. This intermediary sensing modality compensates for the position measurement limitations of pure image processing.
Solution Approach 2:
The system changes the measurement parameters by switching between image-based identification and range-sensor-based positioning. Image processing is used for qualitative parameters (object type, specification) while range sensing provides quantitative parameters (precise distance, position). This parameter change strategy allows the system to leverage the strengths of each sensing modality for different measurement aspects, resolving the contradiction between specification accuracy and position precision.
3Duration of action of stationary object
If range sensor tracking is used in congested environments, then tracking continuity is improved, but reliability deteriorates due to difficulty in specifying object type
Solution Approach 1:
The patent applies preliminary action by using image processing to identify and classify target objects before initiating continuous range sensor tracking. The image processing unit first specifies the object type and characteristics, then this information is used to initialize and guide the subsequent continuous tracking by the range sensor. This preliminary classification ensures that even in congested environments, the system maintains both tracking continuity and object type specification reliability by establishing the target identity before continuous monitoring begins.
Data Source
AI summary
A moving object detecting device measures a congestion degree of a space and utilizes the congestion degree for tracking. In performing the tracking, a direction measured by a laser range sensor is heavily weighted when the congestion degree is low. When the congestion degree is high, a sensor fusion is performed by heavily weighting a direction measured by a image processing on a captured image to obtain a moving object estimating direction, and obtains a distance by the laser range sensor in the moving object estimating direction.