Vehicle Acceleration Prediction from Learned Deceleration Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle systems lack efficient methods to predict and optimize acceleration from learned deceleration areas, which can impact fuel economy, efficiency, and safety.
Innovation Solution
A method and system that utilize sensors and processors to acquire deceleration data, determine if a learned acceleration record exists, and predict vehicle acceleration from a deceleration area based on prior accelerator off speeds, allowing for optimized acceleration to a cruising speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If the vehicle uses traditional acceleration methods without learning from prior deceleration data, then the system complexity is low, but fuel economy and acceleration efficiency are suboptimal
Solution Approach 1:
The system performs preliminary actions by storing acceleration data from previous passages through the same geographic area before the vehicle actually needs to accelerate. This allows the system to have acceleration predictions ready in advance, eliminating the need for complex real-time calculations during the acceleration event itself, thus improving fuel efficiency without proportionally increasing system complexity
Solution Approach 2:
The system creates simplified copies of prior acceleration events by storing key parameters (acceleration rate, duration, speed changes) in a database. These copied data representations allow the system to reference historical performance without replicating the full complexity of original sensor datasets, achieving energy optimization while managing data complexity
2Measurement precision
If the vehicle collects and analyzes detailed deceleration data from multiple sensors, then measurement precision improves, but the device complexity increases
Solution Approach 1:
The system extracts only the essential acceleration parameters (acceleration rate, duration, speed changes, geographic location) from the complete sensor dataset. By taking out only the critical data elements needed for comparison and prediction, the system achieves high measurement precision for the relevant parameters without requiring complex processing of all available sensor information
Solution Approach 2:
The system performs preliminary data filtering and parameter extraction during the deceleration phase when the vehicle is approaching the area of interest. This preliminary processing prepares the data in advance, reducing the computational burden during actual acceleration events and allowing high precision measurements without proportionally increasing real-time system complexity
3Productivity
If the vehicle implements real-time acceleration prediction based on learned deceleration areas, then productivity improves, but the difficulty of detecting and measuring increases
Solution Approach 1:
The system uses copied historical acceleration data stored in a database to predict future acceleration performance. By referencing pre-stored acceleration patterns from previous passages through the same geographic area, the system achieves high productivity without performing complex real-time physics calculations, thereby reducing the difficulty of implementation
Solution Approach 2:
The system implements feedback by comparing predicted acceleration performance with actual measured performance after the acceleration event. This feedback loop allows the system to refine its predictions over time, improving productivity while using relatively simple measurement and comparison logic rather than complex real-time modeling
Data Source
AI summary
Vehicle acceleration from a learned deceleration area or stop can be learned and/or predicted. While approaching a learned deceleration stop, deceleration data including a current accelerator off speed can be acquired. It can be determined whether there is an acceleration learning record for the learned deceleration stop. Responsive to determining that there is an acceleration learning record for the learned deceleration area, it can be determined whether the current accelerator off speed meets a qualification rule based on one or more prior accelerator off speeds included in the acceleration learning record for the learned deceleration area. Responsive to determining that the current accelerator off speed meets the qualification rule, an acceleration for the vehicle from the learned deceleration area can be predicted. The predicted acceleration can include a cruising speed that is substantially equal to the current accelerator off speed.