Event-Camera Forklift Sensing for Dynamic Obstacle Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In dynamic environments like warehouses, existing industrial trucks face challenges in avoiding collisions with dynamic objects due to the lack of effective detection and response systems.
Innovation Solution
Equipping industrial trucks with event cameras that detect local changes in brightness to generate image sensor data, allowing for the determination of object movements and collision probabilities, and using an evaluation device to control the truck's movement or alert the operator through visual, acoustic, or haptic warnings to prevent collisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional cameras are used to detect objects in the environment, then the industrial truck can capture static images, but it cannot effectively detect moving objects or changes in the environment
Solution Approach 1:
The patent changes the detection parameter from static brightness values to dynamic brightness changes over time. Event cameras only trigger when there is a change in brightness, making them inherently sensitive to moving objects while ignoring static scenes. This parameter change enables effective detection of dynamic objects in the warehouse environment.
Solution Approach 2:
The patent replaces conventional frame-based camera systems with event-based cameras that use a fundamentally different detection mechanism. Instead of capturing complete frames at regular intervals, event cameras generate asynchronous events based on local brightness changes, providing more efficient and accurate motion detection.
2Reliability
If the industrial truck is equipped with advanced detection systems to avoid collisions, then safety is improved, but the device complexity increases
Solution Approach 1:
The patent replaces complex conventional vision systems with event cameras that have simpler architectures. Event cameras only generate data when changes occur, reducing overall data volume and processing requirements compared to frame-based cameras that continuously capture complete images regardless of motion.
Solution Approach 2:
By changing from frame-based to event-based detection, the system reduces the amount of data that needs to be processed. Event cameras generate sparse, event-driven data only when motion occurs, simplifying the computational burden on the evaluation device while maintaining high detection reliability.
3Reliability
If the industrial truck continuously monitors the environment for dynamic objects, then collision avoidance is improved, but energy consumption increases
Solution Approach 1:
Event cameras operate in a continuous monitoring mode but only generate output signals when changes occur, effectively creating an event-triggered periodic action pattern. This allows the system to maintain constant environmental awareness while consuming energy only when necessary to process and respond to detected changes.
Solution Approach 2:
The event camera's event-driven operation mode changes the energy consumption pattern from continuous processing of all frames to selective processing only when brightness changes occur. This significantly reduces average energy consumption while maintaining the ability to detect and respond to moving objects in real-time.
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 solution effectively reduces the likelihood of collisions by enabling the truck to detect and respond to dynamic objects in its environment, ensuring safer operation by automatically braking or alerting the operator to potential hazards.
Implementation Method 1
an event camera which comprises an image sensor with a plurality of pixel elements, wherein the plurality of pixel elements are designed to detect local changes in brightness in a field of view of the event camera
Data Source
Figure 1
Figure 2
AI summary
The invention relates to a material handling vehicle (100) for transporting load carriers and/or goods (110) in an environment, in particular a warehouse. The material handling vehicle (100) can be an autonomously, semi-autonomously, and/or manually operated material handling vehicle (100), in particular a forklift truck (100) or industrial robot (100). The material handling vehicle (100) comprises an event camera (120a,b) which includes an image sensor with a plurality of pixel elements, wherein the plurality of pixel elements are configured to detect local changes in brightness within a field of view of the event camera (120a,b) and to generate image sensor data. Furthermore, the material handling vehicle (100) comprises an evaluation unit (130) which is configured to determine object movements of dynamic objects (140) in the environment of the material handling vehicle (100) based on the image sensor data (121). Furthermore, a method for operating such a forklift truck (100) is described.