metadata

license: cc-by-sa-4.0
dataset_info:
  features:
    - name: id
      dtype: int64
  dataset_size: 900000000
task_categories:
  - reinforcement-learning
  - robotics
language:
  - en
annotations_creators:
  - experts-generated
tags:
  - self-driving
  - robotics navigation
pretty_name: FrodoBots (1K)

Dataset Description

Homepage: https://www.frodobots.com/
Hours of tele-operation: ~1,000 Hrs
Dataset Size: 900+ GB
Point of Contact: michael.cho@frodobots.com

FrodoBots 1K Dataset

The FrodoBots 1K Dataset is a diverse collection of camera footage, GPS, IMU, audio recordings & control data collected from close to 1,000 hours of tele-operated sidewalk robot driving in 10+ cities (during internal game testing).

If you're interested in testing out your AI models on your own FrodoBots, we encourage you to support our upcoming Kickstarter Campaign (from US$699 per unit).
If you're interested in testing out your AI models on our existing fleet of FrodoBots (in Singapore, London, Madrid, etc), feel free to contact Michael Cho to gain access to our SDK.
If you're interested in playing the game (ie. remotely driving a FrodoBot), you may sign up here (all you need is a game controller & browser).

We want to express our sincere gratitude to the team at BlurIt for sponsoring our video redaction efforts (ie. blurring of faces & license plates). If you need help with anonymizing your images or videos, feel free to check out their website: https://blurit.io/en


FrodoBots being tested in Stanford University campus

Dataset Summary

There are 7 types of data that are associated with a typical FrodoBots drive, as follows:

Control data: This includes information about the robot's control inputs captured at a frequency of 10Hz.
GPS data: This includes latitude, longitude, and timestamp information for each data point collected during the robot drives at a frequency of 1Hz.
IMU (Inertial Measurement Unit) data: This provides 9-degree of freedom sensor data, including acceleration (captured at 100Hz), gyroscope (captured at 1Hz), and compass ride information (captured at 1Hz), along with timestamp data.
Rear camera video: This offers video footage captured by the robot's rear-facing camera at 20 frames per second (FPS) with a resolution of 480x480.
Front camera video: This consists of video footage captured by the robot's front-facing camera at 20 FPS with a resolution of 800x600.
Microphone: This includes audio recordings captured by the robot's microphone, with a sample rate of 48000Hz.
Speaker: This incorporates audio recordings of the robot's speaker output (ie. gamer's microphone), also with a sample rate of 48000Hz.

In total, there were more than 4,000 individual driving sessions recorded, with majority of these drives under 10 minutes long, although there were also drives lasting more than an hour.


Histogram of Session Duration (minutes)

These drives were done with FrodoBots in over 10 cities, registering close to 1,000 hours of footages, with Liuzhou being the city with longest cumulative hours of drives.


Cumulative sessions hours by city

Out of these 1,000 hours of recorded drive, roughly 312 hours have registered speed above 0 Km/Hr. Most of these other duration were either stationary or short pauses during a regular game drives.


Distribution of robot speed (excluding zero speed)

About FrodoBots

FrodoBots is an open-world video driving game where gamers remotely control sidewalk robots to complete missions in different cities (here's a video about the robot).

The game objective is to complete the pre-defined navigation missions in single-player mode (see video) or gather digital items in the multiplayer arena mode (see video).


Testing in Madrid


Testing in London


Testing in Stockholm


Testing in Wuhan


Testing in Liuzhou


Testing in Berlin


Game Controller + Browser = Control FrodoBots Anywhere


Chatting with locals via built-in microphone/speaker


Zero turning radius = Easy maneuvering


Night driving test in Palo Alto


Driving through rain


Road crossing in Palo Alto

Motivations for open-sourcing the dataset

The team behind FrodoBots is focused on building an open-world video gaming experience using real-life robots (we call it "robotic gaming"). A by-product of gamers playing the game in real-life is the accompanying dataset that's generated.

By sharing this dataset with the research community, we hope to see new innovations that can be tested (via our SDK) directly on our fleet of FrodoBots, and then ultimately deployed back into our production environment in order to create a more fun and safer game.

Helper code

We've provided a helpercode.ipynb file that will hopefully serve as a quick-start for researchers to play around with the dataset.

Contributions

The team at FrodoBots Lab created this dataset, including Michael Cho, Sam Cho, Aaron Tung, Niresh Dravin & Jiamin Ho.