17835035. CONTROL OF VEHICLE AUTOMATED DRIVING OPERATION WITH INDEPENDENT PLANNING MODEL AND COGNITIVE LEARNING MODEL simplified abstract (GM Global Technology Operations LLC)

From WikiPatents
Jump to navigation Jump to search

CONTROL OF VEHICLE AUTOMATED DRIVING OPERATION WITH INDEPENDENT PLANNING MODEL AND COGNITIVE LEARNING MODEL

Organization Name

GM Global Technology Operations LLC

Inventor(s)

Zahy Bnaya of Petach Tikva (IL)

CONTROL OF VEHICLE AUTOMATED DRIVING OPERATION WITH INDEPENDENT PLANNING MODEL AND COGNITIVE LEARNING MODEL - A simplified explanation of the abstract

This abstract first appeared for US patent application 17835035 titled 'CONTROL OF VEHICLE AUTOMATED DRIVING OPERATION WITH INDEPENDENT PLANNING MODEL AND COGNITIVE LEARNING MODEL

Simplified Explanation

The patent application describes a method for controlling automated driving operations using two independent models: a vehicle planning model and a cognitive learning model. These models are connected through a semantic layer, which includes data adaptors.

  • The method involves transforming real traffic data into an abstract representation that can be used by the cognitive learning model to generate a humanized reward model.
  • The vehicle planning model uses the humanized reward model to determine a trajectory plan for the automated driving operation.
  • The automated driving operation is executed by a controller in the vehicle, based on the trajectory plan.

Potential Applications

  • Autonomous vehicles: This technology can be applied to control the operations of autonomous vehicles, improving their planning and decision-making capabilities.
  • Traffic management systems: The method can be used in traffic management systems to optimize traffic flow and reduce congestion.

Problems Solved

  • Lack of human-like decision-making: The cognitive learning model generates a humanized reward model, allowing the automated driving system to make decisions that align with human preferences and behavior.
  • Limited adaptability: The independent models and semantic layer enable the system to adapt to different driving scenarios and learn from real traffic data.

Benefits

  • Improved safety: The method enhances the planning and decision-making capabilities of automated driving systems, leading to safer driving operations.
  • Enhanced efficiency: By using real traffic data and a humanized reward model, the system can optimize driving trajectories and improve overall efficiency.
  • Adaptability: The system can learn from real traffic data and adapt to different driving scenarios, making it more versatile and effective in various environments.


Original Abstract Submitted

A method for controlling an automated driving operation includes setting up respective independent models for executing planning and learning in the automated driving operation, including a vehicle planning model and a cognitive learning model. A semantic layer is generated to act as a bridge between the cognitive learning model and the vehicle planning model, the semantic layer including a first data adaptor and a second data adaptor. The method includes transforming real traffic data to a respective equivalent abstract representation such that it can be used by the cognitive learning model to generate a humanized reward model, via the first data adaptor. The method includes determining a trajectory plan, via the vehicle planning model, based in part on the humanized reward model. The vehicle has a controller that executes the automated driving operation based in part on the trajectory plan.

Retrieved from "http://wikipatents.org/index.php?title=17835035._CONTROL_OF_VEHICLE_AUTOMATED_DRIVING_OPERATION_WITH_INDEPENDENT_PLANNING_MODEL_AND_COGNITIVE_LEARNING_MODEL_simplified_abstract_(GM_Global_Technology_Operations_LLC)&oldid=18204"