Autonomous Vehicle Parking Route Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in learning and navigating to parking slots, especially when obstacles are present or when they need to actively learn multiple parking slots, leading to difficulties in performing autonomous parking.
Innovation Solution
An autonomous vehicle system that determines and stores object properties along a driving route, allowing it to learn and adapt routes, detect empty parking spaces, and park in those spaces by using sensors and user inputs to modify routes and prioritize parking options.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the autonomous vehicle learns a fixed route from starting location to ending location, then the vehicle can navigate autonomously along the learned route, but the vehicle cannot adapt when obstacles are present at the learned parking slot
Solution Approach 1:
The patent implements dynamic route learning where the vehicle continuously updates its route knowledge based on real-time obstacle detection. The system transitions from static pre-learning to dynamic adaptation by modifying the learned route when obstacles are detected during autonomous navigation, allowing the vehicle to adjust its path to alternative parking slots while maintaining autonomous operation.
Solution Approach 2:
The system employs feedback mechanisms where obstacle detection results from sensors are fed back to the route planning module. When obstacles are detected at the learned parking slot, this feedback triggers route recalculation and relearning, enabling the vehicle to adapt its navigation strategy based on environmental conditions while maintaining reliable autonomous parking capability.
2Measurement precision
If the autonomous vehicle detects and stores object properties along the driving route, then the vehicle can identify parking spaces more accurately, but the system complexity increases
Solution Approach 1:
The patent segments the route learning process into distinct phases: initial route learning, object detection during navigation, and parking space identification. By dividing the complex task of autonomous parking into these manageable segments, the system achieves high detection accuracy for parking spaces while keeping each individual module's complexity manageable through modular architecture.
Solution Approach 2:
The system performs preliminary route learning and object property storage before actual parking execution. By pre-storing object properties along the driving route during the learning phase, the vehicle reduces real-time computational complexity during autonomous navigation while maintaining high detection accuracy when identifying available parking spaces.
3Adaptability or versatility
If the autonomous vehicle actively learns multiple parking slots, then the vehicle has more parking options, but the vehicle cannot effectively learn and adapt to various parking scenarios when obstacles are present
Solution Approach 1:
The patent implements self-service through autonomous route recalculation and relearning when obstacles are detected. The vehicle independently identifies alternative parking slots from its learned multiple parking options and autonomously adjusts its navigation path without external intervention, maintaining both versatility in parking slot selection and reliability in successful parking execution.
Solution Approach 2:
The system changes navigation parameters dynamically by switching from a pre-determined learned route to an alternative route when obstacles are detected. By modifying route parameters and selecting different parking slots based on real-time conditions, the vehicle maintains high adaptability across multiple parking scenarios while ensuring reliable parking execution through parameter adjustment rather than system redesign.
Data Source
AI summary
A method of controlling a vehicle, which is configured to be autonomously driven, includes determining a learned route based on a driving route that the vehicle has driven in a manual mode from a starting location to an ending location, driving the vehicle along the learned route in an autonomous mode, detecting a parking space based on driving the vehicle along the learned route in the autonomous mode, and based on a detection of the parking space in the learned route, parking the vehicle in the detected parking space.