Gimbaled Binocular Camera Orientation for Drone Obstacle Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current aerial systems, such as drones, face challenges in achieving omnidirectional depth sensing and obstacle avoidance due to the limited range of stereovision pairs and vulnerability to vibration, especially when rotating, which restricts their ability to detect obstacles in all directions effectively.
Innovation Solution
A system incorporating a pair of wide-angle lens cameras mounted on a gimbal that automatically adjusts its orientation to match the direction of travel, allowing for binocular depth sensing and obstacle avoidance by overlapping fields of view to provide comprehensive depth information and trajectory planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multiple stereovision pairs are stacked to achieve omnidirectional coverage, then the coverage range is improved, but the device complexity and weight increase significantly
Solution Approach 1:
The patent applies a rotatable binocular camera system that can dynamically rotate to different angles (0°, 45°, 90°, 135°, 180°, 225°, 270°, 315°) to achieve omnidirectional coverage. This dynamic rotation mechanism replaces the static approach of stacking multiple camera pairs, reducing device complexity while maintaining full 360° coverage capability.
Solution Approach 2:
A single binocular camera system is designed to perform multiple functions by rotating to different positions. The same camera pair can detect obstacles in all eight directions sequentially, eliminating the need for eight separate camera pairs and significantly reducing system complexity and weight.
2Ease of manufacture
If a fixed binocular camera system is used, then the manufacturing cost is reduced, but the obstacle avoidance capability is limited to specific directions only
Solution Approach 1:
The patent transforms a fixed binocular camera system into a dynamic one by adding a rotation mechanism. The camera can rotate to eight different positions to detect obstacles in all directions, maintaining cost-effectiveness while significantly improving adaptability and obstacle avoidance capability across all orientations.
3Device complexity
If conventional depth sensing methods are used, then the system is simpler, but the measurement precision deteriorates under vibration and angular movement
Solution Approach 1:
The patent employs a rotatable binocular camera system with gimbal stabilization that can dynamically adjust its orientation to match the direction of travel. This dynamic adaptation allows the system to maintain measurement precision under vibration and angular movement by continuously repositioning the cameras to optimal viewing angles, while keeping the overall system relatively simple.
4Area of stationary object
If wide-angle lens cameras are used, then the field of view is improved, but the manufacturing precision requirements increase
Solution Approach 1:
The patent uses wide-angle lens cameras mounted on a rotatable gimbal system. The dynamic rotation capability compensates for the challenging optical characteristics of wide-angle lenses, allowing proper alignment and calibration in different positions. This reduces the stringency of manufacturing precision requirements while maintaining comprehensive field of view coverage.
Data Source
AI summary
An aerial system includes a body, a lift mechanism coupled to the body, a processing system, and at least one camera. The aerial system also includes a first motor configured to rotate the at least one camera about a first axis and a second motor configured to rotate the at least one camera about a second axis. The processing system is configured to determine a direction of travel of the aerial system and to cause the first motor and the second motor to automatically orient the at least one camera about the first axis and the second axis such that the at least one camera automatically faces the direction of travel of the aerial system.