Autonomous Vehicle Stop Planning Using Scored Speed Profiles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous driving systems cannot determine optimal stop locations for autonomous vehicles, often leading to inappropriate stops and failing to consider passenger preferences or road regulations.
Innovation Solution
A method using a speed profile to calculate scores for candidate routes and stop locations, taking into account surrounding information and passenger preferences, to determine the optimal stop location for an autonomous vehicle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing autonomous driving systems use uniform control method for stop, then the control process is simple, but the system cannot consider driver or passenger tendencies and characteristics
Solution Approach 1:
The system dynamically selects from multiple preset speed profiles (first through fourth profiles) based on real-time conditions and user preferences. Each profile represents a different deceleration strategy, allowing the system to adapt its stopping behavior rather than using a fixed uniform control method. This resolves the contradiction by making the control system flexible and adaptable while maintaining manageable complexity through predefined profiles.
Solution Approach 2:
The system changes key control parameters (acceleration, deceleration rates, speed reduction patterns) by selecting different speed profiles. Each profile defines specific parameter ranges for stopping maneuvers, enabling the system to consider different driver/passenger preferences (e.g., smooth vs. quick stopping) without requiring complete redesign of the control architecture.
2Reliability
If existing autonomous driving systems determine stop locations based only on collision prevention, then the control logic is simple, but the vehicle stops at inappropriate locations such as crosswalks or no-stopping zones
Solution Approach 1:
The system performs preliminary evaluation of multiple candidate stop locations before actually stopping. It calculates scores for each candidate location based on multiple criteria (road regulations, safety, suitability) in advance, and selects the optimal location before executing the stopping maneuver. This preliminary scoring and evaluation process ensures reliable stop location selection while managing complexity through structured assessment criteria.
Solution Approach 2:
The system uses feedback from surrounding information (road conditions, traffic rules, detected objects) to evaluate and select stop locations. The scoring mechanism incorporates feedback from multiple sources to determine whether a location is appropriate, ensuring the vehicle avoids crosswalks and no-stopping zones while selecting suitable destinations.
3Measurement precision
If the system calculates scores for multiple candidate routes and stop locations using multiple speed profiles, then the stop location accuracy is improved, but the calculation time and processing complexity increase
Solution Approach 1:
The system segments the stop determination process into distinct stages: generating candidate routes, identifying candidate stop locations on each route, evaluating each candidate using multiple speed profiles, and selecting the optimal option. This segmentation allows the system to systematically evaluate multiple possibilities with high precision while managing computational complexity through structured, modular processing steps.
Data Source
AI summary
Provided are a method, a device, and a computer program for controlling stop of an autonomous vehicle using a speed profile. The method of controlling, by a computing device, stop of an autonomous vehicle using a speed profile includes obtaining surrounding information of an autonomous vehicle, determining candidate routes for controlling stop of the autonomous vehicle on the basis of the surrounding information, calculating scores for candidate driving plans for the autonomous vehicle to travel the determined candidate routes according to a preset speed profile, and finalizing a driving plan for the autonomous vehicle on the basis of the calculated scores.