LIDAR Radar Sensor Fusion for Occluded Object Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current object detection methods using LIDAR and radar sensors face challenges in accurately detecting partially occluded objects due to sensitivity to light changes, limited search range, and processing time, especially at night, and struggle to maintain high detection accuracy.

Innovation Solution

A method and apparatus that combine LIDAR and radar sensors to collect data, extract objects, generate shadow regions, set regions of interest (ROI) based on object movement and reflectivity, and verify object presence by comparing overlapping ROIs, reducing processing time and operation amount.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If LIDAR sensor is used for object detection, then detection accuracy is improved, but processing time increases and detection reliability deteriorates under varying light conditions

Engineering Contradiction:
Improvedetection accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the detection process by dividing the search space into multiple ROIs and processing them in parallel. The processor divides received sensor data into multiple regions of interest and processes each region simultaneously, reducing overall processing time while maintaining detection accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by selectively processing only relevant regions rather than the entire search space. By identifying and processing only ROIs containing potential objects, the system reduces processing time and computational resources while maintaining detection accuracy for critical areas.

Inventive Principle:
Principle #16Partial or excessive action

2Measurement precision

If LIDAR and camera are combined for object detection, then detection accuracy is improved, but device complexity increases and processing time increases

Engineering Contradiction:
Improvedetection accuracyVSAvoidsensor combination complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges LIDAR and radar sensors into a unified detection system that processes data from both sensors simultaneously. The processor integrates data from multiple sensor types to improve detection accuracy while managing system complexity through coordinated processing.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent creates a multi-functional detection system that can handle both LIDAR and radar data through a single processing framework. The processor is designed to universally process data from different sensor types, reducing overall system complexity despite using multiple sensor technologies.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Productivity

If ROI processing is applied to reduce processing time, then productivity is improved, but detection precision may deteriorate for partially occluded objects

Engineering Contradiction:
Improveprocessing speedVSAvoiddetection accuracy for occluded objects
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent performs preliminary actions by pre-identifying and marking ROIs before detailed processing. The system preliminarily scans the search space to identify potential object locations, then focuses detailed processing only on these pre-identified regions, ensuring neither processing speed nor detection accuracy is compromised.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent adds temporal dimension to ROI processing by performing multiple processing passes. The system processes ROIs in different time stages, including initial identification and subsequent verification passes, which improves detection accuracy for occluded objects while maintaining overall processing efficiency.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

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 effectively detects partially occluded objects with reduced processing time and operation amount, improving detection accuracy and robustness, especially at night, by utilizing the strengths of both LIDAR and radar sensors.

Implementation Method 1

a light detection and ranging (LIDAR) sensor may have difficulty recognizing a color of a target to be detected

Methodology Applied
Scientific EffectLight reflection: Reflection

Implementation Method 2

light detection and ranging (LIDAR) sensor

Methodology Applied
Scientific EffectTime of flight: Time of Flight

Implementation Method 3

a radar sensor may be unable to classify a color and a type of a target

Methodology Applied
Scientific EffectRadar reflection: Radar

Implementation Method 4

reflectivity depending on a range and a movement speed of a moving object

Methodology Applied
Scientific EffectDoppler effect: Doppler Effect

Data Source

PatentUS10782399B2Object detecting method and apparatus using light detection and ranging (LIDAR) sensor and radar sensor
Publication Date: 2020.09.22 DAEGU GYEONGBUK INSTITUTE OF SCIENCE AND TECHNOLOGY
  • US10782399B2 patent drawing
  • US10782399B2 patent drawing
  • US10782399B2 patent drawing

AI summary

Provided is an object detecting method and apparatus using a light detection and ranging (LIDAR) sensor and a radar sensor, the method may include collecting LIDAR data and radar data associated with a search region in which objects are to be found using the LIDAR sensor and radar sensor, extracting each of objects present within the search region based on the collected LIDAR data and radar data, generating shadow regions of objects extracted through the LIDAR sensor and setting an ROI of LIDAR sensor based on the extracted objects, setting an ROI of radar sensor based on a reflectivity depending on a range and a movement speed of moving object among the objects extracted through the radar sensor, comparing the ROI of the LIDAR sensor to the ROI of the radar sensor, and verifying whether the moving object is present based on a result of the comparing.