Yield Scenario Encoding for Autonomous Vehicles

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional autonomous vehicles are unable to safely and predictably navigate yield scenarios, such as intersections and merging lanes, due to their inability to encode and deploy traffic regulations and yielding protocols, leading to potential collisions and anxious driving conditions.

Innovation Solution

The system generates a data structure that encodes the geometry of yield scenarios, including contention points and states, which is used by a yield planner to determine appropriate yielding behavior for the autonomous vehicle, ensuring safe and courteous navigation through yield scenarios.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional autonomous vehicles use basic collision avoidance systems, then they can prevent direct collisions, but they fail to encode and deploy traffic regulations and yielding protocols, leading to unpredictable behavior in yield scenarios

Engineering Contradiction:

Improvesafety in yield scenariosVSAvoidability to follow traffic regulations

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent segments the complex task of yield scenario negotiation into distinct components: detecting yield scenarios, encoding geometry data (ego path, contender paths, contention points), determining states of contention, and generating wait elements. This segmentation allows each component to be handled by specialized modules, improving both reliability in yield scenarios and adaptability to different traffic regulations.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary actions by pre-encoding geometry data and determining states of contention before actual yield decisions are made. Wait elements are generated in advance based on encoded geometry and contention states, allowing the autonomous vehicle to prepare yielding behavior proactively rather than reactively, enhancing predictability and safety.

Inventive Principle:

Principle #10Preliminary action

2Ease of operation

If autonomous vehicles fail to properly encode yielding protocols, then system complexity remains low, but other drivers experience anxiety due to unpredictable behavior

Engineering Contradiction:

Improvepredictability of autonomous vehicle behaviorVSAvoidcomplexity of encoding and deploying yielding protocols

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The patent creates simplified copies of human driving decision-making through structured data representations. Geometry data is copied into standardized formats (ego path, contender paths, contention points), and contention states are copied from complex regulatory scenarios into discrete categories. This copying approach enables predictable behavior without requiring full complexity of human-level understanding.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system transforms continuous geometric and regulatory information into discrete parameters and states. Geometry data is converted into structured representations with specific parameters (paths, contention points), and yielding protocols are converted into discrete contention states. This parameterization simplifies processing while maintaining predictability, reducing device complexity.

Inventive Principle:

Principle #35Parameter changes

3Reliability

If conventional systems avoid collisions through basic sensing, then immediate safety is maintained, but long-term safety and efficient negotiation of yield scenarios cannot be achieved

Engineering Contradiction:

Improvelong-term safety through protocol complianceVSAvoiddifficulty of encoding geometry and contention states

Core Design Contradiction:

ReliabilityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent introduces intermediate data structures as mediators between raw sensor data and yielding decisions. Geometry data serves as an intermediary that captures spatial relationships without requiring full scene understanding. Wait elements act as intermediaries that translate contention states into actionable yielding behavior. These intermediaries simplify the detection and measurement tasks while ensuring protocol compliance.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20230130814A1Yield scenario encoding for autonomous systems

Publication Date: 2023.04.27 NVIDIA CORP

AI summary

In examples, autonomous vehicles are enabled to negotiate yield scenarios in a safe and predictable manner. In response to detecting a yield scenario, a wait element data structure is generated that encodes geometries of an ego path, a contender path that includes at least one contention point with the ego path, as well as a state of contention associated with the at least on contention point. Geometry of yield scenario context may also be encoded, such as inside ground of an intersection, entry or exit lines, etc. The wait element data structure is passed to a yield planner of the autonomous vehicle. The yield planner determines a yielding behavior for the autonomous vehicle based at least on the wait element data structure. A control system of the autonomous vehicle may operate the autonomous vehicle in accordance with the yield behavior, such that the autonomous vehicle safely negotiates the yield scenario.