研究目的
To allow a mobile robot vision system to reliably achieve real-time constraints using cloud computing by proposing a human cloud mobile robot architecture and data flow mechanism with specific algorithms.
研究成果
The proposed HCMR architecture and data flow mechanism effectively enhance real-time performance in cloud-based mobile robot vision systems, with improvements in response time, data size, path planning time, and accuracy. Future work should address texture handling, incorporate edge computing, and extend the MGNG algorithm for more geometric surfaces.
研究不足
The mobile robot clustering algorithm does not eliminate redundant clusters that contain textures, which may increase processor utilization. The architecture does not include edge computing technology, and the MGNG algorithm is limited to planar and non-planar surfaces without handling more complex geometric shapes.
1：Experimental Design and Method Selection:
The study involves designing a human cloud mobile robot (HCMR) architecture with a data flow mechanism. Algorithms include a real-time image clustering (RT-IC) algorithm on the mobile robot side and a modified growing neural gas (MGNG) algorithm on the cloud server side.
2：Sample Selection and Data Sources:
Stereo images captured by mobile robots in various environments are used, with different image resolutions and network bandwidths simulated.
3：List of Experimental Equipment and Materials:
Mobile robots equipped with stereo cameras, cloud servers, and simulation tools like CloudSim. Specific hardware includes an Intel
4：5 GHz Corei5 processor with 6 GB RAM. Experimental Procedures and Operational Workflow:
Images are captured, clustered using RT-IC, offloaded to the cloud for processing with MGNG, and results are analyzed for response time, data size, path planning time, and accuracy.
5：Data Analysis Methods:
Response time is calculated using T_t = T_0 + T_p + T_c. Comparisons are made with state-of-the-art techniques using metrics like processing time and accuracy.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
