The dataset is currently empty. Upload or create new data files. Then, you will be able to explore them in the Dataset Viewer.
My Robotics Dataset 11
This dataset contains simple robotics interaction data designed for reinforcement learning and control experiments.
Description
The dataset records state, action, and reward tuples collected from a simulated mobile robot navigating a 2D environment.
Data Format
Each row in the CSV file represents one timestep and contains:
- episode index
- timestep
- robot position (x, y)
- velocity
- action taken
- reward received
- terminal state flag
Intended Use
This dataset can be used for:
- reinforcement learning benchmarks
- robotics control experiments
- navigation policy prototyping
Files
- data.csv: main dataset file
- Downloads last month
- 5