Risk Level Sets for Autonomous Vehicle Path Planning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Autonomous vehicles face challenges in navigating dynamic, congested, and cluttered environments, where existing path planning technologies struggle to effectively quantify congestion and ensure collision avoidance in changing scenarios.

Innovation Solution

The use of risk level sets to quantify the level of risk at different locations in the environment, allowing agents to choose a risk threshold that guarantees collision avoidance by reducing the control space and adapting to the density of obstacles and agents, employing a cost function that maps occupancy to risk and using Djikstra's Algorithm for optimal route planning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional path planning methods are used in congested environments, then the vehicle can navigate through the environment, but the collision avoidance capability is insufficient due to inability to effectively quantify congestion levels

Engineering Contradiction:
Improvecollision avoidance capabilityVSAvoidcongestion quantification complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent transforms the qualitative concept of congestion into a quantitative metric by introducing a cost function that assigns numerical values to different environmental states. This cost function evaluates occupancy, velocity, and acceleration of surrounding agents to generate a scalar congestion metric, enabling the planner to make informed decisions about safe navigation paths without requiring complex explicit congestion models.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If the control space is reduced to ensure collision avoidance, then safety is improved, but the navigation flexibility and ability to adapt to changing environments is limited

Engineering Contradiction:
Improvecollision avoidance guaranteeVSAvoidnavigation adaptability
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The risk level set is constructed dynamically based on current environmental conditions, including the positions, velocities, and accelerations of surrounding agents. As the environment changes, the cost function continuously updates the risk metrics, and the level set adapts to provide collision-free control inputs that are optimized for the current state, enabling both safety guarantees and environmental adaptability.

Inventive Principle:
Principle #15Dynamics

3Reliability

If a conservative risk threshold is chosen to ensure safety, then collision avoidance is improved, but the travel efficiency and speed are reduced

Engineering Contradiction:
Improvecollision avoidanceVSAvoidtravel efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent allows the autonomous vehicle to operate with a chosen risk threshold that balances safety and efficiency. By adjusting the threshold, the system can adopt a conservative approach (lower threshold for higher safety) or a more aggressive approach (higher threshold for better efficiency). This partial action principle enables flexible risk management where the vehicle can tolerate some level of risk in exchange for improved travel efficiency when appropriate.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3794573B1Navigating congested environments with risk level sets
Publication Date: 2023.06.28 MASSACHUSETTS INST OF TECH
  • EP3794573B1 patent drawingFigure 1A
  • EP3794573B1 patent drawingFigure 1B
  • EP3794573B1 patent drawingFigure 2

AI summary

A method is disclosed for use in a planning agent, the method including: identifying a first agent in a vicinity of the planning agent; identifying a location of the first agent and a velocity of the first agent; calculating a set of occupancy costs for the first agent, each occupancy cost in the set of occupancy costs being associated with a different respective location in the vicinity of the planning agent, each occupancy cost in the set of occupancy costs being calculated at least in part based on a cost function that depends on the location of the first agent and the velocity of the first agent; and changing at least one of speed or direction of travel of the planning agent based on the set of occupancy costs.