Intelligent toy tracking trajectory design based on mobile cloud terminal deployment and depth-first search algorithm

View article
Loading...
PeerJ Computer Science

Main article text

 

Introduction

  • 1)

    However, the current approaches are predominantly terminal-based model architectures, which struggle to address the realtime and lightweight requirements essential for smart toys. Furthermore, these existing methods fail to achieve multi-scene trajectory tracking and realtime positioning in unknown environments. To address these limitations, this article proposes a novel smart toy trajectory tracking method called TTNet based on a mobile cloud terminal and depth-first search algorithms. The TTNet method leverages mobile cloud terminals’ computing power and storage capabilities to ensure realtime performance and lightweight implementation. TTNet can maintain rapid response times even for resource-intensive tracking algorithms by offloading complex computations to the cloud. This allows for seamless tracking of smart toy movements, ensuring a smooth and engaging user experience.

  • 2)

    Moreover, TTNet incorporates depth-first search algorithms to enable multi-scene trajectory tracking. By utilizing the search paradigm of depth-first search, TTNet can effectively explore and map unknown environments, enabling accurate realtime positioning even in unexplored scenarios. This capability significantly enhances the adaptability and versatility of the trajectory tracking system, making it suitable for a wide range of real-world applications. Main contributions are as follows: We propose an intelligent toy detection model based on Transformer, which realizes precise positioning of toys by designing an adaptive boundary regression model.

  • 3)

    We propose an intelligent toy trajectory tracking method based on depth-first search, which constructs the inter-frame relationships to achieve realtime tracking.

  • 4)

    Using our approach, we propose a lightweight model deployment method based on mobile cloud terminals to enable mobile applications.

Experiment and analysis

Dataset and implement details

where pr denotes the result from the model, and gt refers to the ground truth. In the context of smart toy tracking, the rationale for selecting mAP and F-value as evaluation metrics lies in their ability to comprehensively and accurately reflect the precision and recall of the model in detecting multiple target toys, thereby effectively assessing the model’s performance.

Compare our detection method with other methods

Compare our tracking method with other methods

Ablation experiments

Conclusion

Supplemental Information

Additional Information and Declarations

Competing Interests

Both authors are employed by Beijing AIQI Technology Co., LTD, The authors declare that they have no competing interests.

Author Contributions

Yang Zhang conceived and designed the experiments, analyzed the data, performed the computation work, prepared figures and/or tables, and approved the final draft.

Hu Zhang performed the experiments, analyzed the data, authored or reviewed drafts of the article, and approved the final draft.

Data Availability

The following information was supplied regarding data availability:

The code is available in the Supplemental File.

The dataset is available at Zenodo: Mavsar, M. (2022). Video-Trajectory Robot Dataset [Data set]. Zenodo. https://doi.org/10.5281/zenodo.6337847.

Funding

The authors received no funding for this work.

538 Visitors 518 Views 18 Downloads

Your institution may have Open Access funds available for qualifying authors. See if you qualify

Publish for free

Comment on Articles or Preprints and we'll waive your author fee
Learn more

Five new journals in Chemistry

Free to publish • Peer-reviewed • From PeerJ
Find out more