Autonomous Vehicle Sensor Data Generation Using Global Coordinate Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In production scenarios, especially individualized production, autonomous vehicles like AGVs face challenges in deterministic behavior due to dynamic components, making it difficult to simulate their interactions and position in virtual commissioning, necessitating sensor data for correct reactions in real environments.
Innovation Solution
A computer-implemented method generates sensor data using an environmental model with a global coordinate system, transforming data into a local coordinate system for individual vehicles, accounting for changing positions and time delays, and transmitting it to controllers for accurate control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If autonomous vehicles use dynamic components for individualized production, then adaptability is improved, but deterministic behavior is lost making simulation difficult
Solution Approach 1:
The patent creates a virtual copy of the production environment including autonomous vehicles, sensors, and environmental models. This virtual replica allows deterministic simulation of dynamic behaviors by copying real sensor data and vehicle states into a controllable virtual space where multiple scenarios can be tested without affecting actual production.
Solution Approach 2:
The patent introduces a sensor data generation system as an intermediary between the physical autonomous vehicles and the simulation environment. This intermediary captures real sensor data from vehicles in the dynamic production environment and transforms it into virtual sensor data, allowing the simulation to bridge the gap between real-world adaptability and virtual determinism.
2Measurement precision
If sensor data is generated for each autonomous vehicle individually, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent merges the sensor data generation processes for multiple autonomous vehicles into a single unified system. The environmental model and sensor configurations are shared across all vehicles, with the system generating virtual sensor data for multiple vehicles simultaneously based on their respective positions and orientations, thereby reducing overall system complexity while maintaining individual measurement precision.
Solution Approach 2:
The patent creates a universal sensor data generation system that serves multiple autonomous vehicles with different sensor configurations. The environmental model and coordinate transformation mechanisms are designed to handle various sensor types and vehicle configurations through a single multi-functional platform, reducing the need for separate generation systems for each vehicle.
3Manufacturing precision
If coordinate transformation is performed for each vehicle, then control accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary coordinate transformations by pre-calculating transformation matrices and environmental model data in the virtual environment before simulation execution. By preparing transformation parameters in advance based on known vehicle positions and orientations, the system reduces real-time processing requirements while maintaining accurate coordinate transformations for control decisions.
Solution Approach 2:
The patent implements dynamic coordinate transformation that adapts to changing vehicle positions and orientations during simulation. Rather than performing static transformations, the system dynamically updates transformation parameters based on current vehicle states, allowing efficient processing that maintains control accuracy as vehicles move through the production environment.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method and a device for generating sensor data for controlling an autonomous vehicle in an environment, such as automated guided vehicles (AGVs) in a factory. Based on an environment model, such as a BIM model, sensor positions of static sensors and the sensors of autonomous vehicles are defined in a global coordinate system. Depending on these sensor positions, sensor data is centrally generated for all sensors in this global coordinate system. The sensor data is then transformed into a local coordinate system of an autonomous vehicle and transmitted for controlling the autonomous vehicle.