Vehicle Task Estimation via Domain Score Accumulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for estimating tasks from a driver's utterance in vehicles are inconvenient and have low accuracy, requiring specific commands and limited operation to prevent erroneous recognition, thus assisting only limited operations.
Innovation Solution
An operating system that estimates tasks by totaling scores based on the importance of domains associated with words from the utterances of both the driver and an operator, selecting the task with the highest total score, and executing relevant actions, such as search or navigation tasks, to improve accuracy and convenience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system accepts only limited command-type utterances to prevent erroneous recognition, then reliability is improved, but ease of operation deteriorates
Solution Approach 1:
The system changes the parameter of utterance acceptance from binary (accept/reject) to continuous scoring. Each domain receives a score based on word importance, and the system accepts the domain with the highest score. This transforms the rigid reliability constraint into a flexible scoring mechanism that maintains reliability while improving ease of operation.
Solution Approach 2:
The system performs self-validation by automatically scoring and selecting domains without requiring explicit command validation. The utterance analysis unit and domain score totaling unit work together to automatically determine the intended domain, reducing the need for restrictive command protocols while maintaining system reliability.
2Measurement precision
If the system requires specific commands with current location, destination, and task details, then measurement precision is improved, but ease of operation deteriorates
Solution Approach 1:
The system changes from requiring complete explicit parameters (current location, destination, task) to using a scoring mechanism that evaluates domain relevance based on keyword matching and word importance. This reduces the burden on the driver while maintaining accurate task estimation through the domain score totaling process.
Solution Approach 2:
The system accepts partial utterance information and uses the domain scoring mechanism to infer the complete task. Instead of requiring the driver to provide all necessary details explicitly, the system processes the utterance and totals domain scores to determine the intended task, reducing the action required from the driver while maintaining measurement precision.
3Reliability
If the system limits accepted utterances to prevent erroneous recognition, then reliability is improved, but adaptability deteriorates
Solution Approach 1:
The system uses a universal domain scoring mechanism that can handle multiple types of utterances (commands, questions, statements) across different domains (navigation, search, settings, etc.). The same scoring process adapts to various utterance types and domains, improving adaptability while maintaining reliability through consistent evaluation criteria.
Solution Approach 2:
The system dynamically adjusts domain scores based on the specific utterance content and word importance. Rather than using fixed acceptance rules, the system calculates scores dynamically for each utterance, allowing it to adapt to different speaking styles and utterance types while maintaining reliability through the scoring mechanism.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An operation determination processing section 207 of a center 200 extracts words included in the utterance of a driver and an operator, reads an attribute associated with each word from a synonym and related word DB 208 in which an attribute is stored so as to be associated with each word, reads a domain of a candidate or the like for the task associated with the attribute from the synonym and related word DB 208 in which domains of a candidate for a task associated with the read attribute or domains of a task to be actually performed are stored, totals the domains read for each word for words included in the utterance of the driver or the like, and estimates those related to a domain with a highest total score as the candidate for the task and the task to be actually performed. In this manner, it is possible to estimate the task with high accuracy.