Autonomous Vehicle Indoor Navigation Using 3D Height Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Semi-autonomous and autonomous motor vehicles face difficulties in navigating structures like parking garages without GPS signals, especially in multi-story buildings with identical aisles, where camera detection relies on legible signs which may be dirty or obstructed.
Innovation Solution
The method employs a digital three-dimensional map of the infrastructure's interior combined with non-optical sensors to orient the vehicle, using fixed infrastructure elements like ventilation pipes and pillars as position markers, allowing navigation even in poor lighting or obscured conditions without additional sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If camera detection is used for navigation in parking garages, then the vehicle can detect signs and markers, but the detection reliability deteriorates when signs are dirty, covered, or illegible
Solution Approach 1:
The patent replaces optical detection (camera-based vision systems) with non-optical sensors such as LIDAR, radar, or ultrasonic sensors. These sensors emit electromagnetic or acoustic waves and detect reflections to create spatial maps of the environment, enabling navigation without relying on visual signs that may be obscured or illegible.
Solution Approach 2:
The patent changes the detection parameter from optical wavelength (visible light) to non-optical wavelengths (electromagnetic waves in microwave or millimeter-wave ranges, or acoustic waves). This parameter change allows the vehicle to penetrate through dust, dirt, and other obstructions that block visible light, maintaining detection reliability in adverse conditions.
2Measurement precision
If GPS signals are used for positioning, then the vehicle can determine its location globally, but positioning becomes impossible in multi-story buildings with identical aisles where GPS signals are unavailable
Solution Approach 1:
The patent introduces an intermediary system consisting of non-optical sensors and onboard processors that create local spatial maps and determine position relative to the vehicle's starting point. This intermediary system bridges the gap by providing positioning functionality in environments where global satellite-based GPS signals cannot penetrate or distinguish between identical structures.
Solution Approach 2:
The patent transitions from two-dimensional map matching (comparing overhead views of aisles) to three-dimensional spatial mapping using depth information from non-optical sensors. By capturing vertical dimension data and structural features at different heights, the system can distinguish between identical aisles on different floors, adding a dimensional cue that GPS alone cannot provide.
3Reliability
If additional sensors are installed in infrastructure to improve navigation, then navigation reliability improves, but infrastructure complexity and cost increase
Solution Approach 1:
The patent enables the vehicle to perform its own environmental mapping and positioning using onboard non-optical sensors, eliminating the need for infrastructure to be equipped with additional sensors. The vehicle independently creates spatial maps and determines its position without requiring external assistance from the building's infrastructure, thereby maintaining infrastructure simplicity.
Solution Approach 2:
The non-optical sensors serve multiple functions: they create spatial maps for navigation, detect obstacles, measure distances to walls and pillars, and identify structural features for position determination. This multi-functionality allows a single sensor system to replace what would otherwise require multiple specialized sensors or infrastructure components, reducing overall system complexity.
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
Enables reliable navigation within buildings by distinguishing between different levels and ignoring parked vehicles, reducing the need for additional infrastructure and improving visibility-independent navigation.
Implementation Method 1
non-optical sensor device (16), in particular a laser sensor, a radar sensor or an ultrasound-based sensor
Implementation Method 2
non-optical sensor device (16), in particular a laser sensor
Implementation Method 3
non-optical sensor device (16), ... or an ultrasound-based sensor
Data Source
AI summary
The disclosure relates to a method for operating a partially autonomous or autonomous motor vehicle. A digital three-dimensional height model of an infrastructure interior which has at least one aisle that can be traversed by the motor vehicle can be provided by a data server device of the infrastructure for example. The height model describes a spatial situation within the infrastructure, and the topography of at least one section of a surface of a region of the infrastructure, said motor vehicle being located in the section, is detected by means of a sensor device of the motor vehicle. A controller of the motor vehicle generates a three-dimensional topographical map of the region by means of the height model using the ascertained topography. The controller ascertains the current position of the motor vehicle within the infrastructure using the result of the comparison, and the controller ascertains a route along the at least one aisle using the ascertained current position and the height model.