Autonomous Vehicle Summon Path Planning Around Parking Lot Obstacles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Human drivers face challenges in efficiently and safely navigating vehicles, particularly in complex environments like crowded parking lots, where existing remote operation technologies are limited to straight-line paths with no intelligence for navigating around obstacles.
Innovation Solution
An autonomous vehicle system that uses sensor data, such as vision data from cameras, to generate a representation of the environment as an occupancy grid, allowing the vehicle to automatically navigate to a specified destination by planning an optimal path based on vehicle operating parameters and continuously updating the environment representation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If remote operation is used to navigate a vehicle from a parking location, then the vehicle can be summoned to a target location, but the route is limited to a single straight-line path with limited steering range and no intelligence in navigating around obstacles
Solution Approach 1:
The patent replaces manual remote control operations with an autonomous driving system that uses sensors, processors, and machine learning models to automatically plan and execute navigation paths. The system substitutes human-operated mechanical steering with automated electronic control that can dynamically adjust to obstacles and environmental conditions.
Solution Approach 2:
The vehicle summoning system performs self-navigation autonomously without requiring continuous human intervention. The autonomous driving system independently processes sensor data, plans paths, controls vehicle actuators, and adapts to obstacles, enabling the vehicle to service itself during the summoning operation.
2Adaptability or versatility
If autonomous navigation with environment representation is implemented, then the vehicle can navigate around obstacles intelligently, but the device complexity increases due to sensors, processing, and machine learning models
Solution Approach 1:
The autonomous driving system is divided into distinct functional modules: sensor data acquisition, environment representation generation, path planning, and vehicle control. This segmentation allows each module to be optimized independently and facilitates systematic integration while managing overall system complexity.
Solution Approach 2:
The autonomous driving system performs multiple functions using a unified architecture: it generates environment representations for both obstacle detection and path planning, processes various sensor types (camera, radar, LIDAR) through a common machine learning framework, and controls multiple vehicle actuators (steering, acceleration, braking) from a single decision-making core.
3Reliability
If the vehicle continuously updates environment representation and replans paths, then navigation accuracy and safety are improved, but the computational load and processing time increase
Solution Approach 1:
The autonomous driving system performs path planning and environment updates at periodic intervals rather than continuously. The processor generates updated environment representations and replans paths at discrete time steps, balancing real-time responsiveness with computational efficiency and energy conservation.
Solution Approach 2:
The system performs preliminary path planning based on predicted trajectories before executing maneuvers. By pre-computing potential paths and evaluating them against predicted environmental conditions, the system reduces the need for frequent emergency replanning, thereby lowering computational load while maintaining safety.
Data Source
AI summary
A processor coupled to memory is configured to receive an identification of a geographical location associated with a target specified by a user remote from a vehicle. A machine learning model is utilized to generate a representation of at least a portion of an environment surrounding the vehicle using sensor data from one or more sensors of the vehicle. At least a portion of a path to a target location corresponding to the received geographical location is calculated using the generated representation of the at least portion of the environment surrounding the vehicle. At least one command is provided to automatically navigate the vehicle based on the determined path and updated sensor data from at least a portion of the one or more sensors of the vehicle.