Radar SLAM Fusion With INS for Real-Time Obstacle Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous moving platforms, such as autonomous cars and UAVs, face challenges in effectively mapping and localizing their environment due to limitations in image processing, including high power consumption, hardware constraints, and poor range estimation, especially under adverse weather conditions and with similar obstacle textures, which restricts their ability to detect and avoid obstacles in a timely manner.
Innovation Solution
A method and system utilizing radar signals for Simultaneous Localization And Mapping (SLAM) that combines radar sensing with Inertial Navigation System data, employing particle filtering and evidence theory fusion to create a point cloud representation of the environment, allowing for accurate separation of real and unreal targets and continuous tracking of objects, even when hidden or out of view.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If image processing methods are used for environment mapping and obstacle detection, then the system can detect obstacles and analyze environment, but the power consumption increases and hardware requirements become more complex
Solution Approach 1:
The patent replaces image processing systems (optical/mechanical) with radar-based detection systems (electromagnetic). The radar system uses electromagnetic wave transmission and reception to detect obstacles and map the environment, substituting the camera-based image processing approach. This substitution reduces power consumption while maintaining or improving detection reliability, especially in adverse weather conditions where image processing fails.
2Measurement precision
If high resolution image processing is implemented, then obstacle detection accuracy improves, but hardware cost and processing power requirements increase significantly
Solution Approach 1:
The patent replaces complex image processing hardware with radar systems that directly measure range through time-of-flight calculations. The radar transmitter sends electromagnetic pulses and the receiver measures the time for echo return, providing direct range measurement without requiring complex optical systems or high-resolution cameras. This substitution achieves accurate range estimation with simpler hardware.
Solution Approach 2:
The patent introduces radar signals as an intermediary medium for environmental sensing. Instead of using light reflection (image processing) or direct distance measurement instruments, the system uses electromagnetic wave propagation and echo timing as an intermediary to indirectly but accurately determine range and map the environment, reducing hardware complexity.
3Measurement precision
If stereoscopic vision is used for range measurement, then distance estimation is possible, but the distance between cameras must be large which is not compatible with small UAV dimensions
Solution Approach 1:
The patent replaces stereoscopic vision (optical system requiring baseline separation) with radar-based time-of-flight measurement (electromagnetic system). The radar system determines range by measuring the time for electromagnetic pulses to travel to and from targets, eliminating the need for physical separation between sensors. This allows accurate range measurement on compact UAV platforms where camera separation would be insufficient.
4Loss of information
If image processing algorithms are used, then environment analysis is possible, but the algorithms are very complex requiring strong processing cores
Solution Approach 1:
The patent replaces complex image processing algorithms with radar signal processing. Instead of analyzing optical images through computationally intensive computer vision algorithms, the system processes radar echo signals to directly extract environmental information such as target range, velocity (via Doppler effect), and spatial distribution. This substitution reduces processing complexity while maintaining information extraction capability.
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 reduces power consumption and hardware requirements, enabling real-time operation with improved obstacle detection and localization, allowing for safer and more efficient navigation of autonomous platforms in various conditions.
Implementation Method 1
a) a radar transceiver for: a.1) transmitting a modulated radar RF signal to an azimuth that corresponds to a desired direction; a.2) receiving reflections of the modulated radar RF signal from real and unreal targets covered by the sector; a.3) outputting an IF signal containing information regarding range, movement direction and speed of each real and unreal targets
Implementation Method 2
receiving data regarding motions parameters of the moving platform from an Inertial Navigation System (INS) module, containing MEMS sensors data
Data Source
AI summary
A method for performing Simultaneous Localization And Mapping (SLAM) of the surroundings of an autonomously controlled moving platform (such as UAV or a vehicle), using radar signals, comprising the following steps: receiving samples of the received IF radar signals, from the DSP; receiving previous map from memory; receiving data regarding motions parameters of the moving platform from an Inertial Navigation System (INS) module, containing MEMS sensors data; grouping points to bodies using a clustering process; merging bodies that are marked by the clustering process as separate bodies, using prior knowledge; for each body, creating a local grid map around the body with a mass function per entry of the grid map; matching between bodies from previous map and the new bodies; calculating the assumed new location of the moving platform for each on the mL particles using previous frame results and new INS data; for each calculated new location of the mL particles with normal distribution, sampling N assumed locations; for each body and each body particle from the previous map and for each location particle, calculating the velocity vector and orientation of the body between previous and current map, with respect to the environment using the body velocity calculated in previous step and sampling N2 particles of the body, or using image registration to create one particle where each particle contains the location and orientation of the body in the new map; for each particle from the collection of particles of the body: propagating the body grid according the chosen particle; calculating the Conflict of the new observed body and propagated body grid, using a fusion function on the parts of the grid that are matched; calculating Extra Conflict as a sum of the mass of occupied new and old grid cells that do not have a match between the grids; calculating particle weight as an inverse weight of the combination between Conflict and Extra conflict; for each N1 location particle, calculating the weight as the sum of the best weight per body for that particle; resampling mL particles for locations according to the location weight; for each body, choosing mB particles from the particles with location in one of the mL particles for the chosen location, and according to the particles weights; for each body and each one of the mB particles calculate the body velocity using motion model; and creating map for next step, with all the chosen particles and the mass function of the grid around each body, for each body particle according to the fusion function.