Dataset Viewer

The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.

Drone-VRP Dataset: Wing Delivery Drone

Generated from RRNCO real-world city road network data with drone-specific constraints.

Drone Specifications

  • Model: Wing Delivery Drone
  • Empty mass: 4.8 kg
  • Battery mass: 1.5 kg
  • Max payload: 2.3 kg
  • Cruise speed: 29.0 m/s
  • Battery capacity: 1080000 J
  • Max flight time: 1200 s

Dataset

  • Instances: 1280
  • Customers per instance: 50
  • Wind: 2.0 m/s
  • No-fly zones per instance: 1
  • Distributions: ['uniform', 'cluster']

NPZ Keys (batch file)

Key Shape Description
locs (B, n+1, 2) Normalized coordinates (depot at index 0)
depot (B, 2) Depot coordinates (normalized)
points_raw (B, n+1, 2) Original lat/lon coordinates
distance_matrix (B, n+1, n+1) Asymmetric road distance (km)
duration_matrix (B, n+1, n+1) Asymmetric travel duration (minutes)
energy_full_matrix (B, n+1, n+1) Energy per edge with full payload (J)
energy_empty_matrix (B, n+1, n+1) Energy per edge with no payload (J)
demand_kg (B, n+1) Demand weight in kg (depot=0)
demand_normalized (B, n+1) Demand as fraction of max payload
payload_capacity_kg (B,) Max payload capacity
battery_capacity_j (B,) Total battery energy budget
max_range_km (B,) Max range with full payload
time_windows (B, n+1, 2) Time windows [start, end] in minutes
service_time (B, n+1) Service time per node (minutes)
speed (B,) Speed in km/min
max_flight_time_s (B,) Max single-flight endurance
wind_vector (B, 2) Wind (x, y) in m/s
wind_strength (B,) Wind magnitude
edge_blocked (B, n+1, n+1) True if edge crosses a no-fly zone
reroute_distance_km (B, n+1, n+1) Detour distance (km) for blocked edges
recharging_stations (B, ...) Indices of recharging station nodes
battery_swap_time_min (B,) Battery swap duration (minutes)
altitude_min_m / altitude_max_m (B,) Altitude constraints
node_type (B, n+1) 0=customer, 1=depot, 2=recharging station

Energy Model

Based on Dorling et al. (2017) momentum-theory power model: P = (m_total * g)^1.5 / sqrt(2 * ρ * A * η)

Energy on leg (i,j): E_ij = P_i * d_ij / v where P_i depends on payload remaining at departure node i.

The model captures the critical drone behavior: range shrinks as payload increases.

Generated by ML Intern

This dataset repository was generated by ML Intern, an agent for machine learning research and development on the Hugging Face Hub.

Usage

from datasets import load_dataset

dataset = load_dataset('aerialblancaservices/drone-vrp-wing')
Downloads last month
6