Introduction
Backgroud & Mission
The Fusion-Oriented Open-Access Data Platform (FOOD) represents a strategic initiative by ISIF and CSIF to advance the field of autonomous systems. Developed by the Smart Sensor Fusion Laboratory. The primary mission of FOOD is to construct a foundational data infrastructure explicitly designed for multi-source information fusion. It aims to resolve the scarcity of high-quality, synchronized, and structurally aligned data required for advanced autonomous systems.
Key Characteristics
To support the evolution of next-generation intelligent agents, FOOD provides a robust verification platform with full-spectrum coverage:
Unified Data Representation: Distinct from traditional datasets that maintain isolated data sturcture for different sensors, FOOD abstracts heterogeneous sensor data into a standardized, fusion-oriented representation. This architectural design breaks down the barriers between modalities, significantly reducing the complexity of multi-source alignment and processing.
Sensor-Agnostic Adaptability: Thanks to this unified design, the dataset possesses inherent universality. It is capable of adapting to virtually all sensor types (Lidar, Radar, Camera, etc.) within a consistent framework. This allows researchers to focus on algorithm design rather than data parsing and format conversion.
Methodological Inclusivity: The unified architecture naturally facilitates fusion across all layers. By standardizing data at the fundamental level, FOOD supports multi-level fusion algorithms more efficiently and intuitively than datasets relying on loose coupling of raw data.
System Architecture
Latest Work