Dataset Description

This page describes the Received Signal Strength Indicator (RSSI) data collected from different Wi-Fi router position in the VisCenter at UNC Charlotte between January 21st and January 25th, 2024. Stickers were placed in alignment creating 9 columns and 11 rows, along with 2 obstacles. Each round of data collection, A Limo Robot was placed in the environment and recorded the strength of the Wi-Fi signal while driving over each of the stickers by human.

Environment Description

Real-world workspace
Fig 1: Photo of the real-world workspace.
SLAM generated map
Fig 2: Map generated by the robot.

The delineation of the workspace is established through the strategic arrangement of distinct materials. After mapping progrss, the robot generates Vis_map.pgm file (as shown in Fig.2) and Vis_map.yaml file. And we provide the mappping file (Python script) to map the position information on the x and y coodination to the pixel information on the pgm file. As depicted in Fig.2, the metallic barrier is constituted by Aluminum Foil, serving as a reflective surface, while the wooden partition is constructed using a robust wood board. The workspace spans a stable area, measuring 5.76 meters in length and 3.84 meters in width.

The workspace can be easily replicated in the real world using the files we have provided, or it can be used to generate a simulated environment for algorithm validation. Researchers can directly adopt our dataset for their studies.

Dataset Overview

The comprehensive dataset is compiled through the diligent data collection efforts of LIMO robots. Each .txt file within the dataset meticulously records the signal strength information gathered by the robots at various calibration points, corresponding to the distinct locations of the router. Every file is systematically named to reflect the specific position of the router during that measurement round. The overall project is structured as follows:

Each router position is shown in this figrue:
Router position
Fig 3: Router position.

Each .txt file is meticulously structured and includes the following elements: x and y coordinates representing the measurement position, x_map and y_map pixels within the PNG file of the generated map, and the corresponding RSSI measurement value. Each .txt data structure is as follows:

x: metes y: metes x_map: pixel y_map: pixel RSSI: db

Visualization

By following the link below, you can interact with the point in Fig.3. Clicking on any point on the map allows you to visualize both the 2D and 3D heatmaps of the RSSI distribution corresponding to different Wi-Fi transmitter positions.

Go to the Heatmap Visualization

Cite Us

Yang, Yupeng, Yiwei Lyu, Yanze Zhang, Ian Gao, and Wenhao Luo. "Integrating Online Learning and Connectivity Maintenance for Communication-Aware Multi-Robot Coordination." 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024, IEEE.