Smart Vehicle Sensor Fusion for Navigation Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current smart vehicle technologies face challenges in efficiently navigating roads, detecting road-side objects, and maintaining safety, particularly in adverse weather conditions, and in managing driver attentiveness and vehicle control.
Innovation Solution
The implementation of a system that uses a combination of cameras, lidar, radar, and sonar sensors for road-side object detection and 3D model creation, along with neural networks for object recognition and vehicle control, and a method to monitor driver attentiveness using camera images and heart rate detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple sensors (cameras, lidar, radar, sonar) are used for road-side object detection and 3D model creation, then navigation accuracy and safety are improved, but device complexity increases
Solution Approach 1:
The patent combines multiple sensor types (cameras, lidar, radar, sonar) into an integrated sensing system that creates a unified 3D model of the environment. This merging approach allows the system to leverage the complementary strengths of each sensor type while managing complexity through integrated processing.
Solution Approach 2:
The sensor system is designed to perform multiple functions simultaneously: object detection, 3D model creation, navigation guidance, and safety monitoring. This multi-functionality reduces the need for separate specialized systems while maintaining high measurement precision across different operational requirements.
2Measurement precision
If neural networks are applied for object recognition and vehicle control, then object detection accuracy is improved, but processing time increases
Solution Approach 1:
The system applies preliminary processing to sensor data before neural network analysis, pre-identifying potential objects and regions of interest. This preliminary action reduces the complexity of the neural network processing required while maintaining high detection accuracy.
Solution Approach 2:
The system implements real-time processing pipelines that skip unnecessary processing steps and prioritize critical detection tasks. Neural networks focus only on the most relevant features and objects, rushing through the processing of less critical data to maintain real-time performance.
3Reliability
If driver attentiveness is monitored using camera images and heart rate detection, then driver safety is improved, but system complexity increases
Solution Approach 1:
The system merges camera-based visual monitoring with physiological sensor (heart rate) detection to create a comprehensive driver attentiveness monitoring system. This combination allows the system to cross-validate signals and improve reliability while managing complexity through integrated processing of multiple data streams.
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
Enhances navigation accuracy and safety by effectively detecting road-side objects and managing vehicle control, even in adverse weather, and ensures driver safety by monitoring attentiveness and taking appropriate actions.
Implementation Method 1
detecting road-side objects with a reflective sensor
Implementation Method 2
capturing images of the road using a camera
Implementation Method 3
applying a trained neural network to detect street signs, cross walks, obstacles, or bike lanes
Implementation Method 4
Vehicles traveling a same route is determined using a vehicle to vehicle communication protocol for identifying peers based upon encoded signals
Data Source
AI summary
Smart car method to navigate a road includes detecting one or more objects using a camera and a sensor to delimit boundaries of a road; creating a 3D model based on outputs of the camera and sensor; and navigating the road with a vehicle.