Checklist ONIA 2026 County Stage, Grades 9–10 · Task 1
Robots in the Maze
Romanian title: Roboți în labirint
Compute four statistics of a robot-experiment dataset and classify each robot's exploration strategy.
The task
A research centre tests autonomous robots in several types of arenas. Each experiment is described by behavioural and environmental measurements (speed, distance, collisions, explored cells, collected items, scan accuracy, coverage ratio and others), and each robot follows one of four strategies: explorer, collector, guardian or sprinter.
Tasks 1–4 ask, on train.csv, for the number of distinct arena types, the maximum average speed, the most frequent arena type and the maximum number of collected items. Task 5 asks for the strategy of every robot in test.csv.
Abridged and translated by SOTA from the official Romanian materials. The official statement has the exact rules, and it wins wherever this summary differs.
At a glance
- You get
train.csv(1,200 rows, 25 columns includingstrategy_label) andtest.csv(400 rows, without the label); a starter kit is attached to the problem. Archive of 248,603 bytes.- You submit
submission.csvwithrobot_id, subtaskID and answer; tasks 1–4 userobot_idGLOBAL, task 5 has one row per test robot.- Scoring
- Tasks 1–4 (15, 15, 10, 10 points): exact match. Task 5 (50 points): Macro-F1 over the four classes; score = 50 × MacroF1 if MacroF1 < 0.93, otherwise 50.
- Rules
- On-site at county competition centres, in a prepared Docker/Jupyter environment (VS Code also available)
- Problem statements unlocked with a password handed out in the room
- Only the listed documentation may be consulted (NumPy, scikit-learn, OpenCV, Python, pandas, Matplotlib, seaborn, PyTorch, SciPy, CatBoost, Joblib)
- Format
- ONIA 2026 county stage, grades 9–10, 14 March 2026 (12:00–16:20 UTC on the platform).