20240034353. AUTOMATIC GENERATION OF CORNER SCENARIOS DATA FOR TUNING AUTONOMOUS VEHICLES simplified abstract (Baidu USA LLC)

From WikiPatents
Jump to navigation Jump to search

AUTOMATIC GENERATION OF CORNER SCENARIOS DATA FOR TUNING AUTONOMOUS VEHICLES

Organization Name

Baidu USA LLC

Inventor(s)

Yu Cao of Sunnyvale CA (US)

Weiman Lin of Sunnyvale CA (US)

Shu Jiang of Sunnyvale CA (US)

Szu Hao Wu of Sunnyvale CA (US)

Jiangtao Hu of Sunnyvale CA (US)

AUTOMATIC GENERATION OF CORNER SCENARIOS DATA FOR TUNING AUTONOMOUS VEHICLES - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240034353 titled 'AUTOMATIC GENERATION OF CORNER SCENARIOS DATA FOR TUNING AUTONOMOUS VEHICLES

Simplified Explanation

The abstract of this patent application describes a method for automatically generating corner simulation scenarios. The method involves generating parameter values, determining their validity based on predefined metrics, and performing simulation tasks to simulate a trajectory planner. A performance score is calculated for each simulation task, and if the score is below a threshold, the parameter values are saved for re-tuning the trajectory planner.

  • The method automatically generates corner simulation scenarios.
  • Parameter values are generated for predefined parameters.
  • The validity of the parameter values is determined based on predefined metrics.
  • Simulation tasks are performed to simulate a trajectory planner.
  • A performance score is calculated for each simulation task.
  • If the performance score is below a threshold, the parameter values are saved for re-tuning the trajectory planner.

Potential applications of this technology:

  • Autonomous driving systems: This technology can be used to automatically generate simulation scenarios for testing and optimizing trajectory planners in autonomous vehicles.
  • Robotics: The method can be applied to generate simulation scenarios for testing and improving trajectory planning algorithms in robotic systems.
  • Virtual reality: This technology can be used to create realistic simulation scenarios for virtual reality applications, such as training simulations for pilots or drivers.

Problems solved by this technology:

  • Manual scenario generation: This method eliminates the need for manual generation of simulation scenarios, saving time and effort.
  • Optimization of trajectory planners: By automatically generating and simulating different scenarios, this technology helps optimize the performance of trajectory planners, leading to improved navigation and control in various applications.

Benefits of this technology:

  • Efficiency: The automatic generation of simulation scenarios and the re-tuning of trajectory planners based on performance scores improve the efficiency of testing and optimization processes.
  • Accuracy: By using predefined metrics to determine the validity of parameter values, this method ensures that only valid scenarios are simulated, leading to more accurate results.
  • Adaptability: The ability to save and reuse parameter values for re-tuning the trajectory planner allows for easy adaptation and improvement of the system over time.


Original Abstract Submitted

embodiments of the invention are provided to automatically generate corner simulation scenarios. in an embodiment, an exemplary method includes performing the following operations for a predetermined number of iterations for each set of predefined parameters. the operations include generating a set of parameter values for the set of predefined parameters; determining whether the set of parameter values is valid or invalid based on a set of predefined metrics; and if the set of parameter values is valid, performing a simulation task to simulate a trajectory planner of the adv in a simulation scenario configured by the set of parameter values. the method further includes calculating a performance score for the simulation task; and if the performance score of the simulation task is below a predetermined threshold, saving the set of parameter values in a storage, wherein the set of parameter values is used for re-tuning the trajectory planner.