ESP32 WiFi CSI Pepper 3D Mapping
Experimental project using ESP32 Channel State Information (WiFi CSI) to attempt non-contact 3D mapping of bell peppers.
Problem
Non-destructive 3D mapping of crops typically requires expensive LiDAR or structured light sensors. WiFi Channel State Information (CSI) offers a low-cost alternative that can detect object presence and potentially reconstruct shapes.
Approach
Using ESP32 microcontrollers with ESP-IDF to capture raw WiFi CSI data. Bell peppers are placed in the signal path between transmitter and receiver. CSI amplitude and phase changes are recorded as synchronized camera frames. Data is processed in Python to extract spatial features.
Data
Raw CSI captures with synchronized camera frames for ground truth. Multiple rotation angles and distances tested.
Validation
In progress. Comparing CSI-derived features against camera-based measurements.
Results
Early stage. Successfully capturing CSI data and camera frames. Processing pipeline under development.
My Role
Sole developer. Hardware setup, firmware development, data collection pipeline.
Next Steps
Complete data collection; develop signal processing pipeline; test ML approaches for shape reconstruction.
Key Outcomes
- ESP32 CSI capture system
- Synchronized camera + CSI data collection
- Experimental 3D reconstruction pipeline