Autonomous Vehicle Path Planning via Adaptive Grid Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous and semi-autonomous vehicle systems face challenges in determining optimal vehicle paths in complex traffic environments, particularly in real-time, due to limitations in simulating the behavior of multiple agents such as autonomous vehicles, non-autonomous vehicles, stationary, and non-stationary objects like pedestrians, which affects smooth operation and collision avoidance.
Innovation Solution
The implementation of a modified Nash equilibrium solution using an adaptive grid search optimization technique based on rewards and penalties for agents, which simulates their behavior to determine optimal vehicle actions and paths, allowing for real-time decision-making at sub-second frequencies to control powertrain, steering, and braking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional path planning methods are used to simulate behavior of multiple agents, then collision avoidance capability is improved, but real-time processing speed deteriorates due to computational complexity
Solution Approach 1:
The patent segments the continuous state space into discrete grid cells, transforming the complex continuous path planning problem into a discrete search problem. This segmentation enables efficient computation by dividing the environment into manageable units that can be processed independently, resolving the contradiction between comprehensive simulation and real-time processing speed.
Solution Approach 2:
The patent pre-computes and stores the utility function values for all possible states in the grid before actual path planning execution. By performing this computation in advance, the system eliminates the need for complex real-time simulations during critical decision-making moments, thereby achieving both reliable collision avoidance and real-time processing speeds.
2Measurement precision
If comprehensive simulation of multiple agent behaviors is performed, then path planning accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent transforms the path planning problem by changing the parameter representation from continuous coordinates to discrete grid cells with associated utility values. This parameter transformation simplifies the computational model while maintaining planning accuracy, as the utility function encapsulates all necessary information about agent behaviors and environmental constraints in a compact form.
Solution Approach 2:
The patent creates a simplified copy of the real environment in the form of a grid representation with utility functions. This abstracted copy captures the essential features needed for path planning without requiring complex simulations of all agent behaviors, thereby reducing computational complexity while preserving planning accuracy.
3Ease of operation
If utility function evaluation is performed at high frequency, then vehicle operation smoothness is improved, but energy consumption increases
Solution Approach 1:
The patent pre-computes utility function values for all grid states before actual vehicle operation. This preliminary action allows the vehicle to query pre-computed values during high-frequency operation updates, achieving smooth control responses without the energy cost of repeated complex calculations, thus resolving the contradiction between operation smoothness and energy consumption.
Data Source
AI summary
A computer, including a processor and a memory, the memory including instructions to be executed by the processor to determine optimal vehicle actions based on a modified version of a Nash equilibrium solution to a multiple agent game, wherein the Nash equilibrium solution is modified by performing an adaptive grid search optimization technique based on calculating rewards and penalties for the agents to determine optimal vehicle actions, wherein the agents include one or more of autonomous vehicles, non-autonomous vehicles, stationary objects, and non-stationary objects including pedestrians and wherein the rewards and the penalties for the agents are determined by simulating behavior of the agents to determine possible future states for the agents to determine the optimal vehicle actions. The instructions can include further instructions to determine a vehicle path based on the optimal vehicle actions and download the vehicle path to the vehicle.