Autonomous Vehicle Occlusion Grid Planning for Safe Intersection Traversal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in navigating through environments with obstacles, as sensors may be blocked by vehicles, buildings, or pedestrians, leading to difficulties in determining safe routes, especially at intersections and with topographic obstacles like hills.
Innovation Solution
The use of an occlusion grid system that combines LIDAR data and image data to determine occluded and unoccluded regions, allowing the vehicle to assess occupancy states and generate safe trajectories by projecting occlusion fields into segmented image data and using ray casting to determine confidence levels in occlusion states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If sensors are used to detect obstacles and plan routes, then navigation capability is improved, but sensor coverage is reduced due to occlusions from vehicles, buildings, and topographic features
Solution Approach 1:
The system performs preliminary actions by predicting potential occlusion regions before the vehicle reaches them, using current sensor data and vehicle trajectory to anticipate where sensors will be blocked, allowing proactive route planning adjustments
Solution Approach 2:
The system transitions from 2D sensor data to 3D spatial reasoning by creating occlusion grids that model three-dimensional occlusion regions caused by vehicles, buildings, and topographic features, enabling comprehensive visibility analysis in multiple dimensions
2Reliability
If occlusion regions are avoided to ensure safety, then collision risk is reduced, but route efficiency decreases due to limited visible areas
Solution Approach 1:
The system dynamically adjusts route planning based on real-time occlusion analysis, continuously updating occlusion grids and reevaluating safe trajectories as the vehicle moves and occlusion conditions change, optimizing both safety and efficiency adaptively
Solution Approach 2:
The system changes parameters by adjusting trajectory confidence levels and safety margins based on occlusion severity, allowing more aggressive routing in areas with low occlusion and more conservative approaches where occlusions are significant
3Measurement precision
If multiple sensors and data processing methods are used to improve occlusion detection, then occupancy state confidence is improved, but computational resources increase
Solution Approach 1:
The system segments the environment into discrete occlusion grids corresponding to different occlusion levels and occupancy states, allowing selective processing of only relevant regions rather than analyzing entire sensor datasets
Solution Approach 2:
The system introduces an intermediary occlusion grid model that translates complex multi-sensor data into simplified occupancy state representations, reducing computational complexity while preserving essential safety information
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the autonomous vehicle's ability to safely traverse complex environments by improving the confidence in occupancy states, reducing the need for excessive resource allocation and minimizing emergency maneuvers, thereby improving safety and user experience.
Implementation Method 1
LIDAR data can be utilized to determine an occlusion state and/or an occupancy state
Implementation Method 2
ray casting the LIDAR data to determine an occlusion state and/or an occupancy state
Data Source
AI summary
Techniques are discussed for controlling a vehicle, such as an autonomous vehicle, based on occluded areas in an environment. An occluded area can represent areas where sensors of the vehicle are unable to sense portions of the environment due to obstruction by another object. An occlusion grid representing the occluded area can be stored as map data or can be dynamically generated. An occlusion grid can include occlusion fields, which represent discrete two- or three-dimensional areas of driveable environment. An occlusion field can indicate an occlusion state and an occupancy state, determined using LIDAR data and/or image data captured by the vehicle. An occupancy state of an occlusion field can be determined by ray casting LIDAR data or by projecting an occlusion field into segmented image data. The vehicle can be controlled to traverse the environment when a sufficient portion of the occlusion grid is visible and unoccupied.