Autonomous Mobile Obstacle Avoidance Using Point Cloud Type Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous mobile devices face inaccuracies in obstacle detection and avoidance due to Lidar errors and obstacle characteristics, leading to insufficient robustness in obstacle avoidance actions.
Innovation Solution
The method involves obtaining point cloud data using a multi-line Lidar device to determine the presence and type of obstacles, employing fast and slow cyclic determinations, and executing specific obstacle avoidance actions based on recognized obstacle types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If Lidar device is used to obtain obstacle information, then obstacle detection capability is provided, but measurement precision deteriorates due to Lidar errors and obstacle characteristics
Solution Approach 1:
The patent combines multiple detection methods (Lidar, vision sensors, ultrasonic sensors, infrared sensors) to form a hybrid detection system. By merging the strengths of different sensing technologies, the system achieves both reliable obstacle detection and precise location measurement, overcoming the limitations of using Lidar alone.
Solution Approach 2:
The patent introduces intermediate processing steps including point cloud data acquisition, obstacle candidate region determination, and multi-sensor data fusion. These intermediary processes act as mediators between raw sensor data and final obstacle detection results, filtering out Lidar errors and improving both reliability and precision through systematic data processing.
2Ease of operation
If obstacle avoidance actions are executed based on Lidar obstacle information, then obstacle avoidance function is provided, but robustness deteriorates due to inaccurate obstacle information
Solution Approach 1:
The patent implements feedback mechanisms where detection results from multiple sensors are continuously compared and validated. The system uses feedback loops to refine obstacle identification, cross-validate Lidar data with other sensor inputs, and adjust obstacle avoidance actions based on confirmed obstacle characteristics, thereby improving robustness while maintaining ease of operation.
Solution Approach 2:
The patent performs preliminary processing of sensor data including point cloud filtering, obstacle candidate region determination, and preliminary obstacle classification before executing avoidance actions. These preliminary actions prepare accurate and validated obstacle information in advance, ensuring that subsequent avoidance actions are based on reliable data, thus improving robustness without complicating the overall operation.
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 enhances the accuracy and robustness of obstacle avoidance actions by accurately identifying and responding to different types of obstacles, improving navigation and operation efficiency.
Implementation Method 1
obtaining point cloud data of the autonomous mobile device in a current work environment
Data Source
AI summary
The present disclosure relates to an autonomous mobile device and its control method and apparatus and a storage medium, the control method includes: an obtaining step configured to obtain point cloud data of the autonomous mobile device in a current work environment; a determination step configured to determine whether there exists an obstacle in the work environment based on the point cloud data; a processing step configured to, when it is determined that there exists an obstacle, recognize a type of the obstacle based on the point cloud data, and execute an obstacle avoidance action corresponding to the type of the obstacle. As such, the robustness of the obstacle avoidance action can be increased.