Simulated Object Instantiation From Log Data for Vehicle Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Creating simulations that accurately reflect real-world scenarios and validate vehicle system functionality is challenging due to discrepancies between simulated and real-world vehicle reactions, particularly with static and dynamic objects, leading to inaccurate simulation results.
Innovation Solution
Techniques for instantiating objects in a simulated environment based on log data by determining the closest location of the simulated vehicle to the real-world vehicle's prior location, updating the simulated environment to include objects perceived at that time, and simulating changes in object appearance with viewpoint changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If simulations are based on data captured by a real-world vehicle, then the simulation can be generated from actual operational data, but the tested vehicle systems react differently than the real-world vehicle, leading to inaccurate validation results
Solution Approach 1:
The patent applies preliminary action by pre-processing real-world log data to create an accurate simulated environment before testing begins. The system reconstructs the real-world environment, objects, and conditions from captured data, establishing a faithful baseline simulation that accurately represents the original operational context before any vehicle system testing is performed.
Solution Approach 2:
The patent implements copying by creating a detailed digital replica of the real-world environment, including static objects, dynamic objects, road geometry, and environmental conditions. This copied simulated environment mirrors the actual physical world conditions from which the log data was collected, enabling accurate comparison between real and simulated vehicle responses.
2Stability of the object's composition
If the simulated vehicle follows the same path as the real vehicle, then the simulation accurately reflects the original scenario, but the tested vehicle systems cannot be properly validated since they need to react to the environment
Solution Approach 1:
The patent applies segmentation by separating the simulation into distinct components: a fixed environmental baseline (road, static objects, weather conditions) and a variable vehicle system under test. This allows the environment to remain stable and consistent with the original real-world scenario while enabling different vehicle controllers and systems to be tested independently within that same environment.
Solution Approach 2:
The patent uses an intermediary approach by introducing a virtual simulated vehicle as a mediator between the static environment and the vehicle systems being tested. This virtual vehicle follows the reference path and interacts with the environment, allowing researchers to observe how different vehicle system implementations respond to the same environmental conditions without altering the environment itself.
3Ease of manufacture
If static objects are represented as stationary in the simulated environment, then the simulation is simpler to create, but dynamic objects and viewpoint changes are not accurately simulated
Solution Approach 1:
The patent applies dynamics by implementing a hybrid object system where static objects (road, buildings, trees) remain fixed in position, while dynamic objects (other vehicles, pedestrians, animals) are given independent motion capabilities. This allows the simulation to accurately represent real-world scenarios where both stationary and moving entities coexist, while maintaining reasonable computational complexity.
Solution Approach 2:
The patent implements local quality by applying different levels of complexity and behavior to different objects in the simulation. Static objects have simple geometric representations, dynamic objects have motion models and behavior rules, and the vehicle under test has full physics and control models. This localized differentiation of object qualities enables accurate simulation of real-world diversity without uniformly high complexity across all elements.
Data Source
AI summary
Techniques for instantiating objects in a simulated environment based on log data are disclosed herein. Some of the techniques may include receiving log data representing an environment in which a real-world vehicle was operating. Using the log data, a simulated environment for testing a simulated vehicle may be generated. The simulated environment may represent the environment in which the real-world vehicle was operating. The techniques may further include determining a location of the simulated vehicle as it traverses the simulated environment. Based at least in part on the log data, a prior location of the real-world vehicle in the environment closest to the location of the simulated vehicle in the simulated environment may be determined. In this way, the simulated environment may be updated to include a simulated object representing an object in the environment that was perceived by the vehicle from the prior location.