Lidar Observation Model for Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately detecting and tracking objects in their surroundings using lidar data, particularly in determining precise object shapes and extents, which affects their decision-making and control maneuvers.
Innovation Solution
A lidar observation model is employed to process lidar data, integrating shape and measurement models to generate accurate shape data structures and update object states, including boundaries, to enhance object detection and tracking precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional lidar processing methods are used to detect and track objects, then the system can operate with simpler processing, but the precision and accuracy of object shape and extent detection is insufficient for effective autonomous decision-making
Solution Approach 1:
The patent segments the object detection process into distinct components: shape data determination from lidar data, extent determination based on shape data, and state updating based on extents. This segmentation allows each component to be optimized independently, improving overall measurement precision while managing system complexity through modular architecture.
Solution Approach 2:
The patent introduces shape data as an intermediary representation between raw lidar data and final object state. This intermediate shape data structure serves as a mediator that captures essential geometric information, enabling more precise extent determination without directly processing complex raw lidar point clouds, thus balancing precision and computational complexity.
2Manufacturing precision
If detailed shape data and multiple extents are determined to improve object boundary accuracy, then the object state and boundary precision are enhanced, but the computational processing time and complexity increase
Solution Approach 1:
The patent performs preliminary determination of shape data from lidar data before determining object extents. This preliminary action creates a structured representation that simplifies subsequent extent calculations, allowing multiple extents to be computed more efficiently from the pre-processed shape data rather than directly from raw lidar points, thus reducing overall processing time while maintaining boundary precision.
3Reliability
If the system processes and interprets detailed environmental information with high precision, then autonomous decision-making accuracy is improved, but the system complexity and computational requirements increase
Solution Approach 1:
The patent applies local quality by determining multiple extents at different levels of detail based on the object state. Rather than uniformly processing all objects with maximum precision, the system adapts the level of extent determination to the specific object and context, improving decision-making reliability for critical objects while reducing unnecessary processing complexity for less critical scenarios.
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
The system enables more accurate and precise object detection and tracking, allowing for effective autonomous vehicle control maneuvers such as nudging, yielding, and braking, while reducing errors and noise in object extent measurements.
Implementation Method 1
one or more lidar sensors configured to output light signals and detect lidar data responsive to receiving return signals corresponding to the outputted light signals
Data Source
AI summary
A system for detecting and tracking objects using lidar can include one or more processors configured to receive lidar data. The one or more processors can determine shape data from the lidar data. The shape data can be indicative of an object. The one or more processors can determine a plurality of extents of the object based on the shape data. The one or more processors can update a state of the object based on the plurality of extents, the state including a boundary of the object. The one or more processors can provide the state of the object to an autonomous vehicle controller to cause the autonomous vehicle controller to control an autonomous vehicle responsive to the state of the object.