Work Vehicle Obstacle Sensing for False Collision Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automatic travel systems for work vehicles face issues with inaccurate obstacle detection due to radar devices, leading to unnecessary collision avoidance operations when tall grass or dust is present, causing erroneous detection of obstacles.
Innovation Solution
An automatic travel system that combines an imaging unit for capturing images and an active sensor for measuring distances, allowing the automatic travel control unit to execute collision avoidance based on accurate separation distance information from the obstacle, thereby reducing false alarms and unnecessary operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a radar device is used as the obstacle detection sensor, then the device complexity is reduced and ease of manufacture is improved, but the measurement precision deteriorates leading to erroneous detection of obstacles
Solution Approach 1:
The patent combines multiple types of sensors (radar device, imaging device, and active sensor such as LiDAR) into an integrated obstacle detection system. The radar provides initial obstacle candidates, the imaging device verifies object identity, and the active sensor measures precise separation distances. This merging of different sensing modalities resolves the contradiction by maintaining manufacturing simplicity while achieving high measurement precision through sensor fusion.
2Reliability
If the peripheral monitoring device warns the driver about obstacles in the warning area, then the reliability of obstacle detection is improved, but unnecessary collision avoidance operations occur when tall grass or dust is present
Solution Approach 1:
The patent introduces an intermediary verification process using an imaging device and active sensor between the radar's obstacle detection and the collision avoidance action. The imaging device acts as an intermediary to verify whether the detected object is a genuine obstacle, and the active sensor provides intermediary distance measurement data. This intermediary verification step filters out false alarms from tall grass or dust while maintaining reliable detection of actual obstacles, thus preventing unnecessary collision avoidance operations.
3Reliability
If the work vehicle performs collision avoidance operations based on radar detection, then the safety is improved, but unnecessary operations are performed reducing productivity
Solution Approach 1:
The patent implements preliminary verification actions before executing collision avoidance operations. The imaging device performs preliminary identification of detected objects, and the active sensor performs preliminary distance measurement verification. Only after these preliminary actions confirm the presence of a genuine obstacle does the system execute collision avoidance operations. This preliminary action approach maintains safety by ensuring real obstacles are detected while preventing unnecessary operations that would reduce productivity.
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 accurately detects obstacles and avoids collisions without performing unnecessary collision avoidance operations, enhancing the reliability and efficiency of automatic travel for work vehicles.
Implementation Method 1
an active sensor that measures a set range in the same predetermined direction as the imaging unit from the work vehicle
Implementation Method 2
a radar device with low object discrimination accuracy is adopted as the obstacle detection sensor
Implementation Method 3
an imaging unit that image-captures a set range in a predetermined direction from the work vehicle
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An automatic travel system for a work vehicle (1) has an automatic travel control unit (23F) that causes a work vehicle, which is provided with an obstacle detection unit (80) that detects obstacles, to travel automatically. The obstacle detection unit (80) includes imaging units (81-84) and active sensors (86-88). When the obstacle detection unit (80) detects an obstacle based on information from the imaging units (81-84), and the separation distance from the obstacle is acquired based on information from the active sensors (86-88), the automatic travel control unit (23F) executes a first collision avoidance control for avoiding a collision between the work vehicle and the obstacle, on the basis of the separation distance from the obstacle acquired based on the information from the active sensors (86-88).