Elevator Pit Safety Net Using LiDAR and Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for a cost-effective and reliable detection system to identify service technicians or mechanics in the elevator pit or on the pit ladder of an elevator system, with minimal maintenance requirements and high detection performance to ensure safety, while minimizing false positives and negatives.
Innovation Solution
A safety net system utilizing a LiDAR sensor or other sensors (such as RADAR or RGBD cameras) arranged in a plane along the bottom of the elevator pit, coupled with a processor that analyzes data to determine the presence of a person using machine-learning algorithms, generating point cloud data to identify hit points indicative of a person's presence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional detection methods are used in the elevator pit, then the system can identify service technicians, but the detection performance is insufficient and false positives/negatives occur frequently
Solution Approach 1:
The patent replaces traditional mechanical or simple sensor-based detection methods with a machine-learning-based detection system. The machine learning algorithm processes data from sensors (such as LiDAR, cameras, or other detection devices) to identify service technicians in the elevator pit with higher accuracy, reducing false positives and negatives by learning patterns from training data.
Solution Approach 2:
The system changes the detection parameters by using multiple sensor types and processing modes. The machine learning model analyzes various parameters from sensor data (position, movement, temporal patterns) to improve detection accuracy. The system can adjust detection sensitivity and processing parameters to optimize performance in different elevator pit conditions.
2Reliability
If a comprehensive detection system is implemented to ensure safety, then detection accuracy improves, but the system complexity and maintenance requirements increase
Solution Approach 1:
The machine learning detection system performs self-improvement by continuously learning from new data and automatically updating its detection models. The system can self-calibrate and adapt to changing conditions in the elevator pit without requiring manual reconfiguration, reducing maintenance complexity while maintaining high safety standards.
Solution Approach 2:
The system incorporates feedback mechanisms where detection results are continuously monitored and used to refine the machine learning models. Feedback from false positives or negatives is fed back into the training process to improve accuracy over time, allowing the system to maintain high reliability while managing complexity through automated optimization.
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 effectively reduces false negative and false positive determinations by distinguishing between valid hit points and background data, ensuring safe operation by preventing the elevator car from entering the pit when a person is present, with the ability to improve accuracy over time through machine-learning.
Implementation Method 1
The sensor is configured to perform sensing to sense an object disposed along the plane and to generate data corresponding to results of the sensing
Implementation Method 2
The sensor is an RGBD camera
Data Source
Figure 1
Figure 2
Figure 3~5
AI summary
A safety net system is provided for an elevator system that includes an elevator pit. The safety net system includes a sensor and a processor. The sensor is arranged in a plane along a bottom of the elevator pit and is configured to perform sensing to sense an object disposed along the plane and to generate data corresponding to results of the sensing. The processor is operably coupled to the sensor and is configured to analyze the data and to determine whether the data is indicative of a person in the elevator pit based on analysis results.