Robot Obstacle Avoidance Using 3D Models and Shortest-Distance Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing obstacle avoidance systems for robots are inadequate as they can only detect obstacles in specific directions and heights, leading to collisions when sensors are not installed in all directions, particularly when moving backwards or encountering obstacles at different heights.
Innovation Solution
A system and method that utilize real-time positioning data and 3D models of robots and obstacles to calculate the shortest distance and movement direction, enabling the robot to avoid obstacles in all directions by preprocessing 3D models into convex forms for accurate collision detection and controlling movement postures accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a proximity sensor is installed on the robot to detect obstacles, then the robot can avoid obstacles in the sensor's detection direction, but the robot cannot detect obstacles in directions where no sensor is installed
Solution Approach 1:
The patent applies universality by using a single type of sensor (laser sensor) to perform multiple detection functions. The sensor is used not only for direct obstacle detection but also for generating point cloud data that is processed into 3D models, enabling the system to detect obstacles in all directions and at various heights without requiring multiple specialized sensors.
Solution Approach 2:
The patent creates a digital copy (3D model) of the physical environment based on laser sensor data. This virtual model allows the system to analyze obstacle positions, heights, and spatial relationships without physical sensors being present in every detection location, effectively extending detection coverage beyond the sensor's physical orientation.
2Device complexity
If the sensor is placed at a fixed height (e.g., 30 cm), then the sensor structure is simple, but the sensor cannot detect obstacles at different heights (e.g., tabletop at 60 cm)
Solution Approach 1:
The patent transitions from 2D planar detection to 3D spatial detection by constructing three-dimensional models of obstacles. This allows the system to detect obstacles at any height by analyzing the Z-coordinate information in the point cloud data, eliminating the limitation of fixed-height sensors while maintaining simple sensor installation.
Solution Approach 2:
The patent changes the detection parameter from fixed height to variable height by processing laser range data into 3D coordinates. The system calculates obstacle height, width, and depth parameters from point cloud data, enabling detection of obstacles at any vertical position without physically adjusting the sensor height.
3Reliability
If multiple sensors are installed in all directions to achieve complete obstacle detection, then the robot can detect obstacles in all directions, but the device complexity and cost increase significantly
Solution Approach 1:
The patent makes the single laser sensor universal by using it for both direct obstacle detection and environmental mapping. The same sensor generates point cloud data that is processed into 3D models, allowing one sensor to perform the work of multiple sensors would otherwise be needed for complete 360-degree coverage.
Solution Approach 2:
The patent introduces a computational intermediary (point cloud processing system and 3D model generation algorithm) that transforms limited sensor data into comprehensive environmental understanding. This virtual processing layer compensates for the physical limitations of having a single sensor, achieving complete obstacle detection without multiple physical sensors.
Data Source
AI summary
A system and method of controlling obstacle avoidance of a robot. The method includes acquiring current positioning data of the robot, and determining whether an obstacle, spaced from a current position at a distance shorter than a preset distance, exists in a path from the current position to a target position or not according to the current positioning data and position data of all obstacles in a predetermined moving region; if the obstacle exists, calculating the shortest distance between the robot and the obstacle according to the acquired positioning data, a predetermined 3D model of the robot and a predetermined 3D model of the obstacle; calculating a due movement direction of the current robot according to the acquired positioning data, the calculated shortest distance and the 3D model of the obstacle, and controlling a movement posture of the robot according to the calculated movement direction to avoid the obstacles.