Robotic Object Detection Using CNN Fusion of Spatial and Infrared Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robotic object detection systems face challenges in accurately identifying and distinguishing multiple types of objects using scanning range sensors, particularly with semi-sparse spatial and infrared data, which limits their ability to determine object type and location effectively.
Innovation Solution
A system and method utilizing a convolutional neural network that processes combined spatial and infrared data from both 2D and 3D sensors to transform sensor data into a multidimensional array, enabling the network to classify and localize objects, including dynamic objects, and predict their trajectories for safe navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional object detection systems are used with scanning range sensors, then the system structure is simple, but the measurement precision of object type and location is insufficient
Solution Approach 1:
The patent replaces traditional mechanical object detection methods with a convolutional neural network (CNN) based computational system. The CNN processes sensor data to automatically identify object types and locations, substituting manual or rule-based detection mechanisms with an intelligent algorithm that learns patterns from training data, thereby significantly improving measurement precision without requiring complex additional hardware
Solution Approach 2:
The patent transforms one-dimensional sensor scan line data into two-dimensional array representations that preserve spatial relationships. This dimensional transformation allows the CNN to process spatial patterns effectively, enabling accurate object type and location detection from what was previously semi-sparse one-dimensional data
2Reliability
If spatial and infrared data are fused using traditional methods, then the processing speed is fast, but the reliability of object detection is insufficient
Solution Approach 1:
The patent merges spatial data and infrared data into a unified two-dimensional array structure where both data types are integrated at the same processing level. The CNN simultaneously processes both data modalities through shared convolutional layers, allowing the system to leverage complementary information from both sensors to improve detection reliability while maintaining computational efficiency through unified processing
Solution Approach 2:
The two-dimensional array representation serves as an intermediary structure that bridges spatial and infrared data. This intermediate representation preserves the spatial relationships from scan lines while incorporating infrared intensity information, enabling the CNN to process both data types together in a standardized format that improves reliability without excessive complexity
3Measurement precision
If high-dimensional relationships in sensor data are maintained, then the measurement precision is high, but the loss of time in processing increases
Solution Approach 1:
The patent performs preliminary organization of sensor data into two-dimensional arrays that preserve spatial relationships before CNN processing. By pre-structuring the data in a format optimized for convolutional operations, the system maintains high-dimensional relationships without incurring excessive processing time during inference, as the transformation is performed efficiently once during data preparation
Solution Approach 2:
The CNN architecture substitutes complex sequential processing with parallel convolutional operations that can process high-dimensional spatial relationships simultaneously. The convolutional layers efficiently extract features from the two-dimensional array representation, maintaining measurement precision while reducing processing time through optimized parallel computation
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
The solution enables robust detection and tracking of various objects, improving navigation safety by accurately determining object types and locations, overcoming limitations of previous systems that struggled with fusing spatial and infrared data and maintaining high-dimensional relationships.
Implementation Method 1
a light detection and ranging (LIDAR) sensor using a convolutional neural network
Implementation Method 2
the 3D sensor further configured to provide infrared data related to one or more of a shape, a size, a type, a reflectivity, a location
Data Source
AI summary
A system includes a mobile robot, the robot comprising a sensor; and a server operably connected to the robot over a network, the robot being configured to detect an object by processing sensor data using a convolutional neural network. A pipeline for robotic object detection using a convolutional neural network includes: a system comprising a mobile robot, the robot comprising a sensor, the system further comprising a server operably connected to the robot over a network, the robot being configured to detect an object by processing sensor data using a pipeline, the pipeline comprising a convolutional neural network, the pipeline configured to perform a data collection step, the pipeline further configured to perform a data transformation step, the pipeline further configured to perform a convolutional neural network step, the pipeline further configured to perform a network output transformation step, the pipeline further configured to perform a results output step.