Autonomous Vehicle Object Detection With Pointwise Property Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting and tracking objects around autonomous vehicles lack sufficient speed, precision, and accuracy, necessitating a more effective approach.
Innovation Solution
A computer-implemented method and system that utilize a machine-learned detector model to process sensor data from autonomous vehicles, including LIDAR data, to classify, cluster, and estimate properties of objects in the vehicle's environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing object detection and tracking methods are used, then the system can identify objects in the environment, but the detection speed, precision, and accuracy are insufficient
Solution Approach 1:
The patent segments the object detection process into distinct stages: sensor data acquisition, point cloud generation, object hypothesis generation, and property estimation. Each stage processes specific data types and produces intermediate results that feed into the next stage, enabling optimized processing at each step while maintaining overall detection accuracy and speed
Solution Approach 2:
The system dynamically adjusts detection parameters including confidence thresholds, search radii, and property estimation weights based on environmental conditions, object types, and sensor data quality. This allows the detection algorithm to optimize its performance characteristics for different operational scenarios, improving both speed and precision adaptively
2Measurement precision
If existing object detection methods are used, then objects can be identified, but the accuracy of object instance property estimations is insufficient
Solution Approach 1:
The patent introduces intermediate data structures including point clouds, object hypotheses, and property estimation models that serve as mediators between raw sensor data and final object properties. These intermediaries enable modular processing where each component can be independently optimized and replaced without affecting the entire system
Solution Approach 2:
The detection system uses universal data structures and processing pipelines that can handle multiple object types (vehicles, pedestrians, cyclists, animals) and multiple property estimations (position, velocity, dimensions, orientation) through a single unified framework, reducing overall system complexity despite the variety of functions performed
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 proposed solution enhances the speed and accuracy of object detection and tracking, improving the safety and efficiency of autonomous vehicle operations by providing more robust object instance property estimations.
Implementation Method 1
a sensor system including at least one LIDAR sensor configured to transmit ranging signals relative to the autonomous vehicle and to generate LIDAR data
Data Source
AI summary
Systems, methods, tangible non-transitory computer-readable media, and devices for detecting objects are provided. For example, the disclosed technology can obtain a representation of sensor data associated with an environment surrounding a vehicle. Further, the sensor data can include sensor data points. A point classification and point property estimation can be determined for each of the sensor data points and a portion of the sensor data points can be clustered into an object instance based on the point classification and point property estimation for each of the sensor data points. A collection of point classifications and point property estimations can be determined for the portion of the sensor data points clustered into the object instance. Furthermore, object instance property estimations for the object instance can be determined based on the collection of point classifications and point property estimations for the portion of the sensor data points clustered into the object instance.