一种基于视觉注意力避障机制的移动机器人行为决策方法
By using a neural network model based on visual attention mechanism and combining it with depth camera detection, a mobile robot was able to quickly and flexibly avoid unexpected obstacles, solving the problems of inflexible response and high computational complexity of existing algorithms when facing uncertain obstacles.
CN117032217BActive Publication Date: 2026-07-17GANTRY LAB
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GANTRY LAB
- Filing Date
- 2023-07-31
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing obstacle avoidance algorithms for mobile robots are not flexible enough when faced with unexpected and uncertain obstacles, resulting in insufficient obstacle avoidance performance and high computational complexity.
Method used
A neural network model based on visual attention mechanism is adopted. Obstacle detection and decision-making are performed through dorsal and ventral attention neural networks. Combined with image information acquired by a depth camera, the biological visual attention mechanism is simulated to achieve rapid obstacle avoidance.
Benefits of technology
It improves the mobile robot's reaction speed and flexibility to unexpected obstacles, reduces computational complexity, and enhances the real-time performance and accuracy of obstacle avoidance.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN117032217B_ABST
Abstract
一种基于视觉注意力避障机制的移动机器人行为决策方法，在每个决策周期中，首先对当前时刻的输入图像进行处理，从图像信息中得到障碍物的空间坐标，然后对当前时刻的障碍物空间坐标与上一时刻的障碍物空间坐标进行比对；当两个时刻的障碍物空间坐标相同时，将障碍物划分为预期不确定障碍物，通过背侧注意力神经网络做出行为决策；当两个时刻的障碍物空间坐标不同时，得出当前时刻的障碍物的碰撞危险度，将其与危险度阈值进行比对，然后通过背侧注意力神经网络或腹侧注意力神经网络做出行为决策。本发明利用视觉信息对注意力网络进行判别，通过神经调节系统实现避障决策，提高了机器人快速应对环境中的意外不确定性障碍物的能力。
Need to check novelty before this filing date? Find Prior Art