Processed Marker-Dotting Data and Analysis Code for HTC-IHE

Description

This dataset contains the processed marker-dotting endpoint-error measurements underlying Figure 8 and Table 2 of the manuscript “HTC-IHE: An In-Hand Estimation Framework Based on Hand–Tactile Calibration for Robotic Manipulation.” The dataset includes 80 valid CAD-based measurements and 80 valid calibrated measurements. Each measurement contains the planar endpoint errors in the x and y directions relative to the fixed target point, expressed in millimetres. The accompanying Python script reproduces the reported centroid bias, area-equivalent radii of the 90% and 80% Gaussian coverage ellipses (r90 and r80), radial-distance standard deviation, and the marker-dotting visualization. This public dataset is limited to the processed marker-dotting data and the corresponding analysis script used for Figure 8 and Table 2.

Steps to reproduce

The dataset contains two CSV files, each comprising 80 valid two-dimensional endpoint-error measurements in millimetres. The first column represents x-direction error and the second column represents y-direction error relative to the fixed target point. To reproduce the reported marker-dotting statistics and Figure 8, use Python 3 with NumPy and Matplotlib installed, then run: python reproduce_marker_dotting.py The script reads cad_valid_80.csv and calibrated_valid_80.csv, verifies the data dimensions and reported statistics, and reproduces the centroid bias, r90, r80, radial-distance standard deviation, and Figure 8. Generated outputs are written to the output directory.

Institutions

• Huazhong University of Science and Technology

  Hubei, Wuhan

Categories

Funders

• National Natural Science Foundation of China

  Beijing, Beijing

  Grant ID: U24A20130
• National Natural Science Foundation of China

  Beijing, Beijing

  Grant ID: 52505016
• Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China

  Grant ID: JYB2025XDXM208

Licence