Version 1.

0
Release Date: April 1st, 2025

6th Kibo Robot

Programming Challenge
Programming Manual

Version 1.0 (Release Date: April 1st, 2025)

Japan Aerospace Exploration Agency (JAXA)

General Point of Contact: [email protected]
Version 1.0
Release Date: April 1st, 2025

List of Changes
All changes to paragraphs, tables, and figures in this document are shown below;

Release Date Revision Paragraph(s) Rationale

April 1st, 2025 1.0 All -
Version 1.0
Release Date: April 1st, 2025

Contents
1. Introduction ............................................................................................................................... 1
2. Setting up your machine ............................................................................................................ 2
2.1. Requirements ..................................................................................................................... 2
2.2. Setting up Android Studio.................................................................................................... 2
3. Creating your application ........................................................................................................... 4
3.1. Creating an Android project................................................................................................. 4
3.2. Writing the application ......................................................................................................... 5
3.3. Building your application ..................................................................................................... 7
4. Running your program on the simulator................................................................................... 12
4.1. Using the simulator server ................................................................................................ 12
4.2. Login ................................................................................................................................. 12
4.3. Uploading the APK and running your program .................................................................. 14
4.4. Checking simulation while running .................................................................................... 15
4.5. Checking the result ........................................................................................................... 15
4.6. Running on your own machine (optional) .......................................................................... 23
5. Programming tips .................................................................................................................... 34
5.1. Do NOT write infinite loops ............................................................................................... 34
5.2. Debugging feature for image processing........................................................................... 35
5.3. Dealing with randomness .................................................................................................. 35
5.4. About navigation errors ..................................................................................................... 37
5.5. Flashlight .......................................................................................................................... 37
5.6. Error handling ................................................................................................................... 37
5.7. Attention to computing resources ...................................................................................... 38
5.8. Performance of Localization .............................................................................................. 39
5.9. Code review ...................................................................................................................... 41
5.10. Setting the application ID ................................................................................................ 41
5.11. Questions and information exchange .............................................................................. 41
5.12. Library for Image Processing .......................................................................................... 41
5.13. Library for machine learning and artificial intelligence ..................................................... 42
5.14. Adding Libraries .............................................................................................................. 42
6. Game API details..................................................................................................................... 44
6.1. Method summary .............................................................................................................. 44
6.2. Method details .................................................................................................................. 45
7. Tips for Astrobee Characteristics ............................................................................................. 51
7.1. Rendering of Astrobee ...................................................................................................... 51
7.2. Specification of Astrobee................................................................................................... 53
7.3. References of Astrobee .................................................................................................... 53
 Version 1.0
Release Date: April 1st, 2025

1. Introduction
Let's start programming!
Astrobee can be controlled with an Android application called the Guest Science APK
(GS APK). First, setup your machine to build your application according to the instructions
in Chapter 2. Next, read Chapter 3 and create a GS APK. Then, try running the GS APK in
the simulator environment. Chapter 4 describes how to use the simulator environment.

1
 Version 1.0
Release Date: April 1st, 2025

2. Setting up your machine

First of all, set up a machine for programming.

2.1. Requirements
The machine must meet the following requirements.

 64-bit processor
 4 GB RAM (8 GB RAM recommended)
 Ubuntu 20.04 (64-bit version) ( http://releases.ubuntu.com/20.04/ )
or Windows 10 (64-bit version) or Windows 11 (64-bit version)

NOTE: If you want to run your program on your own PC, you need 16 GB of RAMand
Ubuntu 20.04. For details, please refer to 4.6. Running on your own machine (optional).

2.2. Setting up Android Studio

2.2.1. Installing components (Only on Ubuntu)
If you use an Ubuntu machine, you need these components.

 openJDK8
 ADB (Android Debug Bridge)
 Gradle

Please install them with the following command.

sudo apt-get -y install openjdk-8-jdk adb gradle

2.2.2. Installing Android Studio

Please download Android Studio 3.6.3 from the Android Studio download archives page
( https://developer.android.com/studio/archive) and extract it to your home directory.

2
 Version 1.0
Release Date: April 1st, 2025

2.2.3. Downloading additional components

To build the Guest Science APK, you need to download additional components as
follows.

1) Launch Android Studio.

2) Select [Tools] -> [SDK Manager].
3) On the SDK Platforms tab, check “Show Package Details” and select “Android SDK
Platform 25”, “Android SDK Platform 26”, and “Android SDK Platform 28”.
4) On the SDK Tools Tab, check “Show Package Details” and select “25.0.3”, “26.0.2”,
“28.0.3” under Android SDK Build-Tools and select “20.0.5594570” under NDK (Side
by side).”
5) Click the [Apply] button to install these components.

3
 Version 1.0
Release Date: April 1st, 2025

3. Creating your application

3.1. Creating an Android project
To create your application, prepare a new project with the following steps.

1) Download the APK (Sample APK or Template APK) from the download page on the
web site.
2) Extract the zip file to the directory where you want it.
3) Launch Android Studio.
4) Open the APK template folder with [File] -> [Open].
5) Open [app/java/jp.jaxa.iss.kibo.rpc.sampleapk/YourService.java] or
[app/java/jp.jaxa.iss.kibo.rpc.defaultapk/YourService.java] in Project view.
6) Write your code in runPlan1 – runPlan3 methods in the YourService.java file.

When you open the APK template folder, the “Android Gradle Plugin Update
Recommended” dialog may appear. However, you must not update the plugin because
of a dependency problem, so select “Don’t remind me again for this project.”

Figure 3-1 Android Gradle Plugin Update Recommended dialog

4
 Version 1.0
Release Date: April 1st, 2025

3.2. Writing the application

You can use the game APIs shown below in the YourService.java file.
“runPlan1” is executed on the web simulator. You can choose any plan when you run
the application on your own machine.

public class YourService extends KiboRpcService {

@Override
protected void runPlan1(){
// The mission starts.
api.startMission();

// Move to a point.
Point point = new Point(10.9d, -9.92284d, 5.195d);
Quaternion quaternion = new Quaternion(0f, 0f, -0.707f, 0.707f);
api.moveTo(point, quaternion, false);

// Get a camera image.

Mat image = api.getMatNavCam();

/* *********************************************************************** */
/* Write your code to recognize type and number of landmark items in the each area! */
/* If there is a treasure item, remember it. */

/* *********************************************************************** */

// When you recognize landmark items, let’s set the type and number.
api.setAreaInfo(1, "item_name", 1);

/* **************************************************** */
/* Let's move to the each area and recognize the items. */
/* **************************************************** */

// When you move to the front of the astronaut, report the rounding completion.
point = new Point(11.143d, -6.7607d, 4.9654d);
quaternion = new Quaternion(0f, 0f, 0.707f, 0.707f);
api.moveTo(point, quaternion, false);
api.reportRoundingCompletion();

/* ********************************************************** */
/* Write your code to recognize which target item the astronaut has. */
/* ********************************************************** */

5
Version 1.0
Release Date: April 1st, 2025

// Let's notify the astronaut when you recognize it.

api.notifyRecognitionItem();

/*
******************************************************************************************************* */
/* Write your code to move Astrobee to the location of the target item (what the astronaut
is looking for) */
/*
******************************************************************************************************* */

// Take a snapshot of the target item.

api.takeTargetItemSnapshot();
}
protected void runPlan2() {
// You can write your other plan here, but it’s not run on the web simulator.
// …
}
}

You can find methods of the game APIs by using the code completion function of
Android Studio.
Please refer to 6. Game API details for more information and you can download a
sample APK from the download page on the web site.

6
 Version 1.0
Release Date: April 1st, 2025

3.3. Building your application

3.3.1. On Ubuntu
To build your application, use the command shown below.
NOTE: DO NOT build your application using Android Studio to change the build task, as
this may cause an error.

$ cd <YOUR_APK_PATH>
$ ANDROID_HOME=$HOME/Android/Sdk ./gradlew assembleDebug

You can find the APK file here: “<YOUR_APK_PATH>/app/build/outputs/apk/debug/app-

debug.apk”.

3.3.2. On Windows
Please build your application with the following steps.

1) Launch the Android Studio.

2) Open <YOUR_APK_PATH>.
3) Click app on the [Project] window.
4) Select [Build] -> [Make Module ‘app’].
If you find errors, please build an APK file on an Ubuntu machine.

You can find the APK file here: “<YOUR_APK_PATH>\app\build\outputs\apk\debug\app-

debug.apk”.

3.3.3. Change the application ID, APK name and short name (optional)
You can change the application ID (jp.jaxa.iss.kibo.rpc.sampleapk or
jp.jaxa.iss.kibo.rpc.defaultapk by default).
In this step, we change the application ID to “jp.jaxa.iss.kibo.rpc.myteam” and the APK
name/short name to “myteam” with the SampleAPK project.
NOTE: This instruction is for the Final Round. Changing the application ID is not
necessary and not recommended in the Preliminary Round.
NOTE: “jp.jaxa.iss.kibo.rpc” cannot be changed.

1) Launch Android Studio.

2) Open <YOUR_APK_PATH>.
3) Make sure you are viewing the project in Android View.

7
Version 1.0
Release Date: April 1st, 2025

Figure 3-2 Android View

4) Click on settings (the gear icon) and deselect [Compact Middle Package].

Figure 3-3 Unselect [Compact Middle Package]

5) Please expand the “java” folder.

Figure 3-4 Expand the “java” folder

6) Right-click the “sampleapk” folder and select [refactor] -> [rename].

8
Version 1.0
Release Date: April 1st, 2025

7) A warning will be displayed, but you want to go ahead and click [Rename Package].
After that, enter theAPKname that you want. (In the picture, we rename as it
“myteam”.)

Figure 3-5 Rename dialog

8) At the bottom of Android Studio, “Refactoring Preview” will be displayed. Here, click
[Do Refactor].

Figure 3-6 Refactoring Preview

9
Version 1.0
Release Date: April 1st, 2025

9) Open build.gradle (Module: app) in Gradle Scripts on the left-side of the menu.
Please change the application ID and click [Sync Now].

Figure 3-7 build.gradle (Module: app)

10) Open strings.xml in app -> src -> main -> res -> values on the left-side of the
menu. Please change the APK name and save it.

Figure 3-8 strings.xml

11) Open commands.xml in app -> src-> main -> res -> xml on the left-side of the
menu. Please change the value of “shortName” tag and save it. If there is no
“shortName” tag, please add it under the “apkInfo” tag and set your short name.
<shortName>myteam</shortName>

Figure 3-9 commands.xml

You have successfully changed the application ID in Android Studio. If you want to
change the Android project name and its directory name, follow the following steps.

10
Version 1.0
Release Date: April 1st, 2025

◼ On Ubuntu
12) Close Android Studio.
13) Please execute the following commands.

cd <YOUR_APK_PATH>
cd ../
mv SampleApk <YOUR_APK_NAME>
cd <YOUR_APK_NAME>
mv SampleApk.iml <YOUR_APK_NAME>.iml

◼ On Windows
12) Close the Android Studio.
13) Please rename a SampleApk folder to <YOUR_APK_NAME> with Windows Explorer.
14) Now rename SampleApk.iml to <YOUR_APK_NAME>.iml in the SampleApk folder with
Windows Explorer.

11
 Version 1.0
Release Date: April 1st, 2025

4. Running your program on the simulator

4.1. Using the simulator server
Once you have built your application, you can run it on the web simulator provided by
JAXA. To use the simulator, you need a user account issued by the Kibo-RPC secretariat. If
you don’t have one, please read the Kibo-RPC Guidebook to complete your application for
participating in Kibo-RPC first.
Note that the actual Kibo environment is not exactly the same as the simulation
environment, since there are many objects in the Kibo and the environment changes
frequently. Please refer to the image and Google Map for the actual environment. (These
are not the latest.)

Figure 4-1 Image inside Kibo

Google Map: https://www.google.com/maps/@29.5604024,-
95.0855631,2a,75y,205.79h,103.61t/data=!3m6!1e1!3m4!1sUA46_vIbk9kAAAQvxgbyMg!2
e0!7i10000!8i5000

Also, the 12stro of Astrobee’s NavCam and DockCam are different between the
simulator and the real robot. Refer to the image below.

Figure 4-2 (Left) Actual snapshot (Right) Simulator image

4.2. Login
Access the Kibo-RPC web site ( https://jaxa.krpc.jp/ ) and click “LOGIN.”

12
Version 1.0
Release Date: April 1st, 2025

Figure 4-3 LOGIN tab

On the login page, enter the ID and password for your team’s account, and click the
“LOGIN” button. If you have forgotten your ID, please contact the Kibo-RPC secretariat.
You can reset your password by clicking the “Forgot your Password?” link.

Figure 4-4 LOGIN button

On the home page, click “SIMULATION.”

Figure 4-5 SIMULATION button

13
 Version 1.0
Release Date: April 1st, 2025

Now, you can access the web simulator from this page.

Figure 4-6 Web simulator page

4.3. Uploading the APK and running your program

On the simulation page, there are three slots for simulations, so that you can run at most
three programs in parallel.
To start your simulation, select your APK file, the simulator version, and the simulation
conditions.
Enter a memo if desired, and click the “START SIMULATION” button.

Figure 4-7 START SIMULATION button

A simulation may take longer than 20 minutes to run, and it does not need your attention
while it runs. After starting your simulation, you can log out, get a cup of coffee, then go
back to the web site.

When there is a simulation running, the slot displays its original information, and you
cannot run another simulation in the same slot until it finishes.
If you want to stop your simulation, click the “TERMINATE SIMULATION” button. Note
that terminating a simulation loses its game score and output files (such as rosbag and the
Android Emulator’s log).

14
 Version 1.0
Release Date: April 1st, 2025

Figure 4-8 TERMINATE SIMULATION button

4.4. Checking simulation while running

When your simulation is running, you can log in to the simulator server (viewer) via your
browser. Click the “SIMULATIOR VIEWER” button to show the information for a remote
connection and open the viewer in another tab by clicking the “VIEW” button.

Figure 4-9 VIEW button

Enter the password for your remote connection to log in. Now you can use rviz to see
how Astrobee moves in your simulation. This viewer is available until the simulation is
finished.
The viewer displays a real-time simulation in the view-only mode for the simulation
stability. You cannot operate the viewer.

4.5. Checking the result

4.5.1. Result summary
Once your simulation has started, you can check the results by clicking the “VIEW
RESULT” button on the simulation page.

Figure 4-10 VIEW RESULT button

On the result page, you can see the details of your simulation, such as the game time,
the matching of areas and items, and so on.

15
Version 1.0
Release Date: April 1st, 2025

Figure 4-11 RESULT page

Table 4-1 Report Result

Astrobee found the lost item the astronaut was looking for
Green check
Found the at the correct position within the time limit.
correct Astrobee executed takeTargetItemSnapshot API
item Red cross at a wrong position. Or Astrobee did not try to execute
takeTargetItemSnapshot API.
*Refer to the rulebook for the defined correct position.

Click “Matching of Area and Item Detail”, detail of mapping information is displayed.

16
Version 1.0
Release Date: April 1st, 2025

Figure 4-12 Matching of Area and Item Detail

Table 4-2 Target Result

Green check The type and number of Items reported were correct.
Recognition
Red cross The type and number of Items reported were wrong.

17
 Version 1.0
Release Date: April 1st, 2025

4.5.2. Download ZIP file

You can get a ZIP file by clicking the “DOWNLOAD LOG FILES” button. This ZIP file
contains the game score and the Android Emulator’s log. The output log level is INFO.
Note that some or all of these files will not be available unless your simulation finishes
properly. Besides the result page, the game score also appears in a JSON file, which can
be read using a text editor.
Table 4-3 Example of result.json
{
“Mission Time”: {
“start”: “20240123 203208003”, “Mission Time” is the difference between the
“finish”: “20240123 203456003” “start” time and the “finish” time.
},
“RoundingReport”: {
“timestamp”: “20240123 203456003”,
“arrival”: {
“try”: true, “try” is true if you execute
reportRoundingCompletionAPI. “success” is true
“success”: true,
if you execute reportRoundingCompletionAPI at
“distance”: 3.2 the correct position.
},
“areaMapping”: {
“areas”: [
{
“areaId”: 1,
“level”: 1,
“correct”: true, “correct” is true if both “lost_item” and “num”
“lost_item”: “item_name”, are correct.
“num”: 1 The string for “lost_item” needs to match the
}, string described in the rulebook.

{
“areaId”: 2,
“level”: 3,
“correct”: false,
“lost_item”: “ item_name”,
“num”: 1
},
{
“areaId”: 3,
“level”: 2,
“correct”: true,
“lost_item”: “ item_name”,
“num”: 1
},
{

18
Version 1.0
Release Date: April 1st, 2025

“areaId”: 4,
“level”: 4,
“correct”: true,
“lost_item”: “ item_name”,
“num”: 1
}
]
}
},
“FoundReport”: { “try” is true if you execute
“timestamp”: “20210315 203323123”, takeTargetItemSnapshotAPI.

“cameraAngle”: 45, “success” is true if you execute

takeTargetItemSnapshotAPI at the correct
“distance”: 3.2, position.
“try”: true,
“isInArea”: true,
“success”: true
},
“Illegal”: false, “Illegal” is for server internal use.
"OasisZoneN": 300,
"TimeBonusN":10

You can also get a rosbag as a ZIP file by clicking the “DOWNLOAD ROSBAG FILE”
button. The size of this file will be so large that it may take a long time to download it.
In addition, you can save any debug images with saveBitmapImage/saveMatImage API
and download the images by clicking the “DOWNLOAD IMAGE FILES” button.

19
 Version 1.0
Release Date: April 1st, 2025

4.5.3. Check simulation after running

To check previous simulations, click the “Results” button on the simulation page. The
results page lists your past simulations. This list can hold up to 20 simulations.

Figure 4-13 Results list page

The “VIEW RESULT” button is the same as the one on the simulation page. Please be
careful when you click the “REMOVE RESULT” button; it removes the output files of the
selected simulation and these results will be lost.
You can play the rosbag (simulation result) at 0.5x – 3x speed with the viewer. You can
change rosbag replay settings and rviz settings. The details are, described in the sections
below.

20
 Version 1.0
Release Date: April 1st, 2025

4.5.4. rosbag replay settings

You can change rosbag replay settings using Rosbag Player.

Type Description

Speed Select replay speed.

Slider

Range Select replay range.

Selector

Play Start replay and open rviz window.

Button If rviz already has opened, it will
restart.

Pause Pause replay.

Button

Resume Resume replay.

Button

Stop Stop replay and back to

Button PlayConfig window.

Figure 4-14 Rosbag Player

21
 Version 1.0
Release Date: April 1st, 2025

4.5.5. rviz settings

You can change the display settings for the rviz window.

Table 4-4 rviz configuration

Item Check box in the “Displays” tab

Planning trajectory [Visualize]->[PlanningTrajectory]

Trajectory [Visualize]->[Trajectory]

KeepInZone/KeepOutZone [Visualize]->[Zones]

NavCam [Sensors]->[NavCam]

DockCam [Sensors]->[DockCam]

HazCam [Sensors]->[HazCam]

Figure 4-15 rviz configuration description

22
 Version 1.0
Release Date: April 1st, 2025

4.6. Running on your own machine (optional)

You can also run the program on your own machine. This chapter provides a procedure
to set up the Astrobee Simulator. You get a simple simulation environment without
randomness modules (Phases/targets randomness generator, air flow simulator and
navigation errors) or a judge module.

4.6.1. Differences between web simulator and local simulator

Local Simulation Environment do not include random factor modules (object type and
position, airflow, and navigation error).
You can test and debug your program using a local simulator, but you need to evaluate it
on a web simulator server in order to obtain a high score in the Preliminary Round.

4.6.2. Requirements
The following requirements are needed to set up a simulation environment on your
machine.

 64-bit processor
 16 GB RAM
 Ubuntu 20.04 (64-bit version) ( http://releases.ubuntu.com/20.04/ )
 Disk: 30GB of free space (SSD is highly recommended)

4.6.3. Overview
The overall procedure is as follows.
1. Installing Docker
2. Building the Kibo-RPC Simulator with Docker
3. Setting Android Emulator to run APK
4. Building the Guest Science Manager APK
5. Setting up the network to connect the Android emulator and the simulator
6. Installing your APK to Android Emulator
7. Launching the simulator and run your APK
i. Launching the Android Emulator
ii.Starting the Kibo-RPC Simulator
iii.Running the Guest Science Manager, GDS Simulator and your APK

4.6.4. Installing Docker

Please install Docker engine before setting the simulator.

23
 Version 1.0
Release Date: April 1st, 2025

$ sudo apt-get update

$ sudo apt-get install ca-certificates curl gnupg lsb-release
$ curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo gpg --dearmor -o
/usr/share/keyrings/docker-archive-keyring.gpg
$ echo \
“deb [arch=$(dpkg –print-architecture) signed-by=/usr/share/keyrings/docker-archive-
keyring.gpg] https://download.docker.com/linux/ubuntu \
$(lsb_release -cs) stable” | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null
$ sudo apt-get update
$ sudo apt-get install docker-ce docker-ce-cli containerd.io
$ sudo groupadd docker
$ sudo usermod -aG docker $USER

Then, logout abd log back in.

For more details, see official instructions:
https://docs.docker.com/engine/install/ubuntu/
https://docs.docker.com/engine/install/linux-postinstall/

4.6.5. Building the Kibo-RPC Simulator with Docker

Download the Kibo-RPC Simulator Setting up scripts from our website.

( https://jaxa.krpc.jp/download.html )
This module contains the following scripts.
➢ build.sh
➢ run.sh

The build.sh clones the Astrobee Robot software from GitHub, applies patches of Kibo-
RPC modules and builds docker images.

All you have to do is to execute the script as follows.

Note that the building sequence will take several hours to complete and needs large
disk space (20-30 GB).
※Builds may fail under certain conditions. When the build fails, check the CPU and
memory utilization, delete unnecessary processes, and re-run the build.

unzip krpc6_simulator.zip -d $HOME

cd ${HOME}/local_simulator
docker load -i krpc_astrobee_sim_local.tar

Once building the docker images is finished, you can launch the simulator.

24
Version 1.0
Release Date: April 1st, 2025

Edit hosts file to set up the network.

sudo nano /etc/hosts

Add the following three lines and save the file.

127.0.0.1 hlp
127.0.0.1 mlp
127.0.0.1 llp

If hlp, mlp, and llp settings already exist, comment them out, add three new lines, and
save them.

#10.42.0.36 hlp
#10.42.0.35 mlp
#10.42.0.34 llp

127.0.0.1 hlp
127.0.0.1 mlp
127.0.0.1 llp

Finally, you can run a simulation.

bash run.sh

Is the image below displayed on your screen? If so, installation is complete!

Figure 4-16 Setup result

25
Version 1.0
Release Date: April 1st, 2025

The following error will appear, but this is because the device is not connected. Please
ignore this error as the simulator will work normally.

[ERROR] [1742302049.035362566, 7.632000000] : (ros.depth_odometry)

image_features_with_known_correspondences_aligner_depth_odometry.cc:87:
DepthImageCallback: Time difference too large, time diff: 1.336

Type Ctrl+C on the terminal to stop the simulator, and proceed to the next step.

26
 Version 1.0
Release Date: April 1st, 2025

4.6.6. Setting the Android Emulator to run APK

Create an AVD ( Android Virtual Device ) as follows.

1) Launch Android Studio.

2) Select [Tools] -> [AVDManager].
3) In the Android Virtual Device Manager window, click [+ Create Virtual Device …].
4) Select device Nexus 5 ( Resolution 1080x1920 ) and click [Next].
5) Select the [x86 Images] tab, choose Nougat/API Level 25/ABI x86_64/Android
7.1.1(NO Google APIs), then click [Next].
NOTE: Download the system image now if you need it.
6) Set the AVD name to “AstrobeeAndroidSim.”
7) Click [Finish].

In the Android Virtual Device Manager window, you will see “AstrobeeAndroidSim” in the
list.
Click the Play button in the Action column. If the AVD launches successfully, you will see
the following image.

Figure 4-17 Android Emulator screen

See https://github.com/nasa/astrobee_android/blob/master/emulator.md for details.

Shutdown the emulator and let’s procced to the next step.

27
 Version 1.0
Release Date: April 1st, 2025

4.6.7. Building the Guest Science Manager APK

To run your program, you must install the Guest Science Manager APK. (see details at
https://github.com/nasa/astrobee_android/blob/master/guest_science_readme.md )

At first, clone the Astrobee Android package from GitHub.

export ANDROID_PATH=${HOME}/astrobee_android
git clone https://github.com/nasa/astrobee_android.git
git checkout a8560ab0270ac281d8eadeb48645f4224582985e

Next, build the Guest Science Manager APK.

Modify the Guest Science Manager build.gradle as follows:
 Modify the Guest Science Manager build.gradle as follows.

vi $ANDROID_PATH/core_apks/guest_science_manager/build.gradle

28
Version 1.0
Release Date: April 1st, 2025

 Replace the contents of build.gradle with the following.

// Top-level build file where you can add configuration options common
to all sub-projects/modules.

buildscript {
repositories {
mavenCentral()
}
dependencies {
classpath 'com.android.tools.build:gradle:2.3.0'

// NOTE: Do not place your application dependencies here; they

belong
// in the individual module build.gradle files
}
}

allprojects {
repositories {
maven {
url
"https://github.com/rosjava/rosjava_mvn_repo/raw/master"
}
maven {
url
"https://raw.githubusercontent.com/rosjava/rosjava_mvn_repo/master/"
}
maven {
url "https://artifacts.camunda.com/artifactory/public/"
}
mavenLocal()
mavenCentral()
}
}

task clean(type: Delete) {

delete rootProject.buildDir
}

 Run the build.

cd $ANDROID_PATH/core_apks/guest_science_manager

29
 Version 1.0
Release Date: April 1st, 2025

ANDROID_HOME=$HOME/Android/Sdk ./gradlew assembleDebug

4.6.8. Setting up the network

Setup the network between the Kibo-RPC Simulator and the Android Emulator.
See https://github.com/nasa/astrobee_android/blob/master/emulator.md for details.

◼ Setting the HOST network

Execute the following command to open the hosts file.

sudo nano /etc/hosts

Comment out the previous three lines, add new three lines below and save the hosts
file.
#127.0.0.1 hlp
#127.0.0.1 mlp
#127.0.0.1 llp

10.42.0.36 hlp
10.42.0.35 mlp
10.42.0.34 llp

◼ Setting the environment variables

Execute the following commands to set the environment variables.

export ANDROID_PATH=$HOME/astrobee_android
export EMULATOR=$HOME/Android/Sdk/tools/emulator
export AVD=”AstrobeeAndroidSim”

Note that you need to execute the above commands whenever you open a terminal. If
you write these commands in your .bashrc file, you don’t have to execute them.

◼ Setting up the Android network and starting the Android Emulator

Execute the following commands to set up the Android network and launch the Android
Emulator.

30
 Version 1.0
Release Date: April 1st, 2025

cd $ANDROID_PATH/scripts
./launch_emulator.sh -n

See https://github.com/nasa/astrobee_android/blob/master/emulator.md for details.

* If the Android network does not work, try turning off Wi-Fi in the Android Emulator.

4.6.9. Installing APKs

If the Android Emulator is not running, execute the following commands to start it.

cd $ANDROID_PATH/scripts
./launch_emulator.sh -n

In another terminal, execute the following commands to install the Guest Science
Manager APK and your GS APK.

cd $ANDROID_PATH/core_apks/guest_science_manager
adb install -g -r activity/build/outputs/apk/activity-debug.apk
cd <YOUR_APK_PATH>
adb install -g -r app/build/outputs/apk/debug/app-debug.apk

4.6.10. Running your program

It's time to run your program!

◼ Launching the Android Emulator

Execute the following commands to launch the Android Emulator if it is not running.

cd $ANDROID_PATH/scripts
./launch_emulator.sh -n

◼ Starting the Kibo-RPC Simulator

Execute the following command to start the Kibo-RPC Simulator.

cd ${HOME}/local_simulator
bash run.sh

31
Version 1.0
Release Date: April 1st, 2025

◼ Running the Guest Science Manager, GDS Simulator and your GS APK
Execute the following commands to start the Guest Science Manager APK and to
launch the GDS simulator.
Note: It’s required to re-execute these commands whenever you re-launch the simulator
or re-install your APK.

ANDROID_PATH/scripts/gs_manager.sh start

docker exec -it astrobee bash

cd /src/astrobee/src/tools/gds_helper/src
python3 gds_simulator.py

Operate the GDS simulator to run your GS APK.

1) Press any key to grab control.

2) Select your GS APK.
3) Type b and press Enter to start the GS APK.
4) Type d and press Enter to send a custom guest science command.

Now Astrobee starts to locate the item!

※ When the GDS Simulator times out

The following steps may solve the problem.

◼ Launching the Android Emulator

Execute the following commands to launch the Android Emulator.

cd $ANDROID_PATH/scripts
./launch_emulator.sh

◼ Starting the Kibo-RPC Simulator

Execute the following command to start the Kibo-RPC Simulator.

cd ${HOME}/local_simulator
bash run.sh

◼ Running the Guest Science Manager, GDS Simulator and your GS APK

32
Version 1.0
Release Date: April 1st, 2025

Execute the following commands to start the Guest Science Manager APK and to
launch the GDS simulator.

adb shell su 0 sh /cache/emulator_setup_net.sh

bash $ANDROID_PATH/scripts/gs_manager.sh start
docker exec -it astrobee bash
cd /src/astrobee/src/tools/gds_helper/src
python3 gds_simulator.py

33
 Version 1.0
Release Date: April 1st, 2025

5. Programming tips
5.1. Do NOT write infinite loops
You must not write any infinite loops in your code because no one can stop Astrobee
while the loop is executing.
Double check that you use finite loops with a defined counter value, as shown below.

// NG
while(!result.hasSucceeded()){
// do something
}

// OK
final int LOOP_MAX = 5;
int loopCounter = 0;
while(!result.hasSucceeded() && loopCounter < LOOP_MAX){
// do something
++loopCounter;
}

34
 Version 1.0
Release Date: April 1st, 2025

5.2. Debugging feature for image processing

You can save any Bitmap/Mat type images in the Android Emulator and download the
images from the dashboard display. This feature should be useful to check intermediate
images of your image processing algorithm.
To save an image, use saveBitmapImage or saveMatImage API as follows.

// get/process a bitmap image

Bitmap image = any_function();
// save the image
api.saveBitmapImage(image, “file_name_1”);

// get/process a mat image

Mat img = any_function_mat();
// save the image
api.saveMatImage(img, “file_name_2”);

If you are running APK on your local machine, the image can be obtained by the
following command.

(If your APK is based on TemplateAPK)

adb pull /sdcard/data/jp.jaxa.iss.kibo.rpc.defaultapk/immediate/DebugImages

(If your APK is based on Sample APK)

adb pull /sdcard/data/jp.jaxa.iss.kibo.rpc.sampleapk/immediate/DebugImages

Up to 50 images can be saved per simulation, and the maximum image size is 1228800
pixels (1280 x 960 px).

5.3. Dealing with randomness

You must consider the randomness of the environment.
When you want to move the robot, refer to the commands below…

// move to point 1
api.moveTo(point1, quaternion1, true);
// move to point 2
api.moveTo(point2, quaternion2, true);
// move to point 3
api.moveTo(point3, quaternion3, true);

If there is no randomness in the environment, this code works well.

35
Version 1.0
Release Date: April 1st, 2025

However, Astrobee may be faced with errors such as tolerance violations and
collision detection (*), and your code will not work, so you have to provide redundant
code, as we see below.
Remember, Do NOT allow any infinite loops in your code!
* Tolerance violation error occurs when there is a discrepancy between Astrobee’s pose
(estimated) and the target pose. Collision detection occurs when Astrobee’s HazCam
detects any obstacles on the target path. Both errors can occur for a variety of causes
including false detection, especially in the real environment.

Result result;
final int LOOP_MAX = 5;

// move to point 1(first try)

result = api.moveTo(point1, quaternion1, true);

// check result and loop while moveTo api is not succeeded.

// Do NOT write infinite loop.
int loopCounter = 0;
while(!result.hasSucceeded() && loopCounter < LOOP_MAX){

// retry
result = api.moveTo(point1, quaternion1, true);
++loopCounter;

}
// move to point 2
//…

36
 Version 1.0
Release Date: April 1st, 2025

5.4. About navigation errors

The real world always has uncertainties.Navigation error is one of them and the Kibo-
RPC simulator server simulates it.
However, modeling and simulating navigation errors are highly complicated, and this
increases the calculation load. Therefore, random error following Gaussian distribution is
used generally.
The Kibo-RPC simulator also implements a Gaussian distribution and the parameters
are as follows;
Regarding position;
x: mean = 0 m and 3sigma = 0.1 m
y: mean = 0 m and 3sigma = 0.1 m
z: mean = 0 m and 3sigma = 0.1 m
Regarding orientation;
x: mean = 0 degree and 3sigma = 3 degree
y: mean = 0 degree and 3sigma = 3 degree
z: mean = 0 degree and 3sigma = 3 degree

You have to consider that self-positioning and self-orientation obtained from the APIs
(getRobotKinematics) includes these errors.

5.5. Flashlight
You must consider the effect of light conditions on the image recognition in the real
environment. In our past experiments, when the brightness value was set to 0.05 or higher
and taken a snapshot in front of the image, the light reflected on the illustration and the
image could not be recognized properly.
// turn on the front flashlight
api.flashlightControlFront(0.01f);
// you might need some sleep…
Thread.sleep(2000);
// get a camera image
Mat image = api.getMatNavCam();
// turn off the front flashlight
api.flashlightControlFront(0.0f);

5.6. Error handling

Various errors can occur due to external and internal causes during program run.
Considering all kinds of situations and handling the errors make your program more robust
to environmental changes.

Null check: Ensure variable is not null.

Mat image = api.getMatNavCam();

37
 Version 1.0
Release Date: April 1st, 2025

// You must handle null pointer

If (image == null) {
/** Error handling**/
} else {
readAR(image);
}

Size of list/array: Check the length of list/array when access its element.
List<String> list = getStringList();

// Bad
for (int i=0; i<5; i++) {
// this code cause an exception when the list size is less than 5.
String str = list.get(i);
}

// Good
for (int i=0; i<list.size(); i++) {
String str = list.get(i);
}

try-catch: Handle exceptions with try-catch statements.

try {
String decodedStr = someDecoder();
int value = Integer.parseInt(decodedStr);
// …
} catch (NumberFormatException e) {
/** Handle the exception occurs when the string is not number **/
} catch (Exception e) {
/** Handle unexpected exceptions to prevent program down **/
}

5.7. Attention to computing resources

If the computing loads are high, Astrobee might be overloaded and not work on orbit.
The specifications of Astrobee real robot’s HLP are as follows. Note that available
resources are different by other software working on HLP.
Multithreading is not recommended, but in our past experiments, up to two simultaneous
threads can work without problems.
CPU: Qualcomm Snapdragon 820 (4 cores, 2.2GHz)
RAM: 4GB

38
 Version 1.0
Release Date: April 1st, 2025

5.8. Performance of Localization

There is a possibility of losing Astrobee’s self-position on-orbit. Once the self-position is
lost, Astrobee may not be able to recover on its own. In this case, it means it is a game
over. Be aware that this incident is not in the simulator. It is well known that the technology
Astrobee uses is likely to lose its self-position when the navigation camera view gets too
close to wall, floor, airlock, and so on because the camera cannot capture enough features
at those places. Please note the above when creating your program.

For more information, mapped landmark (ML) features coverage heat map is shown in
the figures below.
The localization performance depends on how many ML features NavCam captures.
The heat map represents the number of mapped landmark (ML) features in a volume of 30
㎤ along an imaginary grid on the overhead, aft, forward, deck walls and the front of the
airlock.
The more ML features in a given volume the more stable Astrobee localization will be.

Overhead Wall Airlock Entrance

Fwd

Aft x
y

Figure 5-1 ML features on overhead wall

Aft Wall Airlock Entrance

y
Deck
z
Figure 5-2 ML features on aft wall

39
Version 1.0
Release Date: April 1st, 2025

Forward Wall Airlock Entrance

Deck
y
z
Figure 5-3 ML features on fwd wall

Deck Wall Airlock

Fwd Entrance

Aft
x
y
Figure 5-4 ML features on deck wall

Airlock

x
z
Figure 5-5 ML features on Airlock

40
 Version 1.0
Release Date: April 1st, 2025

5.9. Code review

In the Final Round, before the submitted APK is uplinked to Astrobee on orbit, JAXA /
ARC performs a code review for safety confirmation in advance. In the code review, if there
is an inappropriate code in the submitted APK, we might delete it or instruct the participants
to rewrite it.
To ensure stability, should we confirm that the code may utilize more than three threads
simultaneously, we will instruct the participants to rewrite it.
5.10. Setting the application ID
Each Final Round APK must have a unique application ID to avoid conflict when
installing on Astrobee in the ISS. The application ID will be specified for the finalists later.

5.11. Questions and information exchange

You can post questions, share programming tips and exchange information with other
teams on the bulletin board. Make effective use of it to create your program!

Figure 5-6 BULLETIN BOARD page

5.12. Library for Image Processing
We recommend you to use OpenCV as the image processing library. The version
'opencv-contrib:4.5.3.0' was used for our verification of normal functioning. We do not
guarantee the other versions.

◼ Pattern Matching Notes

When utilizing pattern matching with OpenCV, it’s important to be mindful of certain
considerations.
 The matchTemplate function executes without changing the size of the input template
image. You will need to implement image resizing yourself.
 The matchTemplate function operates without changing the angle of the input
template image. You will need to implement image rotation yourself.

41
 Version 1.0
Release Date: April 1st, 2025

 Depending on the threshold of similarity, there’s a possibility that the same area of
the image may be detected multiple times.
 If the target image for pattern matching is large, it can result in longer processing
times. Let’s consider source codes to reduce processing time.
 The matchTemplate function measures similarity by taking into acount the blank
areas of the template image. You may want to crop the template image for use.

5.13. Library for machine learning and artificial intelligence

We recommend you to use TensorFlow Lite if you use AI for your image processing.
https://www.tensorflow.org/lite

The version ' tensorflow-lite-task-vision:0.4.0' was used for our verification of nominal
functioning. We do not guarantee the other versions.
https://mvnrepository.com/artifact/org.tensorflow/tensorflow-lite-task-vision/0.4.0

5.14. Adding Libraries

If there is a library you wish to use in your program, it is possible to add it to the source
code.
To add a library to the source code, please insert the necessary entries in the
dependencies section of the <YOUR_APK_PATH>/app/build.gradle file.
dependencies {
compile project(':guest_science_library')
compile project(':kibo_rpc_api')
compile fileTree(dir: 'libs', include: ['*.jar'])
compile 'com.android.support:appcompat-v7:26.+'
compile 'com.android.support.constraint:constraint-layout:1.0.2'
compile('org.ros.rosjava_core:rosjava:[0.2,0.3)') {
exclude group: 'xml-apis'
exclude group: 'junit'
}
compile('org.ros.android_core:android_10:[0.2,0.3)') {
exclude group: 'xml-apis'
exclude group: 'junit'
}
testCompile 'junit:junit:4.12'
compile 'com.quickbirdstudios:opencv-contrib:4.5.3.0'

// Additional Library
compile ' org.tensorflow:tensorflow-lite-task-vision:0.4.0'

42
Version 1.0
Release Date: April 1st, 2025

※The source code provided by the Kibo-RPC secretariat is set to a specific version of the
build tool. This version cannot be changed. When adding libraries, please verify in
advance whether they are compatible with the currently set version of the build tool. If
there is no compatibility, the library may not be usable.

43
 Version 1.0
Release Date: April 1st, 2025
6.
6. Game API details
Details of the Kibo-RPC’s game APIs are listed below.

6.1. Method summary

Table 6-1 Method summary

Modifier and Type Method and Description

gov.nasa.arc.astrobee.Resul flashlightControlBack(float brightness)

t
Set the brightness of the back flash light.

gov.nasa.arc.astrobee.Resul flashlightControlFront(float brightness)

t
Set the brightness of the front flash light.

Bitmap getBitmapDockCam()
Gets Bitmap image of DockCam.

Bitmap getBitmapNavCam()
Get Bitmap image of NavCam.

double[][] getDockCamIntrinsics()
Get camera matrix and distortion coefficients of
DockCam.

org.opencv.core.Mat getMatDockCam()
Gets Mat image of DockCam.

org.opencv.core.Mat getMatNavCam()
Gets Mat image of NavCam.

double[][] getNavCamIntrinsics()
Get camera matrix and distortion coefficients of
NavCam.

gov.nasa.arc.astrobee.Kinem getRobotKinematics()
atics
Gets current data related to positioning and orientation
for Astrobee.

gov.nasa.arc.astrobee.Resul moveTo(gov.nasa.arc.astrobee.types.Point
t goalPoint,
gov.nasa.arc.astrobee.types.Quaternion or
ientation, boolean printRobotPosition)
Moves Astrobee to the given point and rotates it to the
given orientation.

44
 Version 1.0
Release Date: April 1st, 2025

Modifier and Type Method and Description

void notifyRecognitionItem()
Notify that Astrobee has recognized the TargetItem held
by the astronaut.

gov.nasa.arc.astrobee.Resul relativeMoveTo(gov.nasa.arc.astrobee.types.
t Point goalPoint,
gov.nasa.arc.astrobee.types.Quaternion or
ientation, boolean printRobotPosition)
Moves Astrobee to the given point using a relative
reference and rotates it to the given orientation.

boolean reportRoundingCompletion()
Report rounding completion and blink the lights.

void saveBitmapImage(Bitmap image,

java.lang.String imageName)
Save a bitmap image for debug.

void saveMatImage(org.opencv.core.Mat image,

java.lang.String imageName)
Save a mat image for debug.

void setAreaInfo(int areaId,

java.lang.String itemName)
Set matching of an area and an item.

void setAreaInfo(int areaId,

java.lang.String itemName, int number)
Set matching of an area and an item, and the number of
an item.

void shutdownFactory()
This method shutdown the robot factory in order to
allow java to close correctly.

boolean startMission()
Undock Astrobee from docking station.

void takeTargetItemSnapshot()
Take a snapshot of TargetItem. When execute this
method, the mission is completed.

6.2. Method details

• getRobotKinematics

45
 Version 1.0
Release Date: April 1st, 2025

public Kinematics getRobotKinematics()

Gets current data related to positioning and orientation for Astrobee. Note that the data cannot be
trusted when the confidence is POOR or LOST.
Returns:
Current Kinematics.

• getBitmapNavCam

public Bitmap getBitmapNavCam()

Gets Bitmap image of NavCam.

Returns:
Bitmap image of NavCam(1280 px x 960 px) or null if an internal error
occurs. Format:Bitmap.Config.ARGB_8888

• getBitmapDockCam

public Bitmap getBitmapDockCam()

Gets Bitmap image of DockCam.

Returns:
Bitmap image of DockCam(1280 px x 960 px) or null if an internal error
occurs. Format:Bitmap.Config.ARGB_8888

• getMatNavCam

public Mat getMatNavCam()

Gets Mat image of NavCam.

Returns:
Mat image of NavCam(1280 px x 960 px) or null if an internal error
occurs. Format:CV8UC1

• getMatDockCam

public Mat getMatDockCam()

Gets Mat image of DockCam.

Returns:
Mat image of DockCam(1280 px x 960 px) or null if an internal error
occurs. Format:CV8UC1

• flashlightControlFront

public Result flashlightControlFront(float brightness)

Set the brightness of the front flash light.

In the final round, brightness greater than 0.01f will be reset to 0.01f.
Parameters:
brightness - Brightness percentage between 0 - 1.
Returns:

46
 Version 1.0
Release Date: April 1st, 2025

A Result instance carrying data related to the execution. Returns null

if the command is NOT executed because of an error

• flashlightControlBack

public Result flashlightControlBack(float brightness)

Set the brightness of the back flash light.

In the final round, brightness greater than 0.01f will be reset to 0.01f.
Parameters:
brightness - Brightness percentage between 0 - 1.
Returns:
A Result instance carrying data related to the execution. Returns null
if the command is NOT executed because of an error.

• moveTo

public Result moveTo(Point goalPoint,

Quaternion orientation,
boolean printRobotPosition)

Moves Astrobee to the given point and rotates it to the given orientation.
Parameters:
goalPoint - Absolute cardinal point (xyz)
orientation - An instance of the Quaternion class. You may want to use
CENTER_US_LAB or CENTER_JEM as an example depending on your initial
position.
printRobotPosition - Flag whether to print robot positions in log or
not.
Returns:
A Result instance carrying data related to the execution. Returns null
if the command is NOT executed because of an error.

• relativeMoveTo

public Result relativeMoveTo(Point goalPoint,

Quaternion orientation,
boolean printRobotPosition)
Moves Astrobee to the given point using a relative reference and rotates it to the given orientation. This
api can take up to 30 seconds when localization is not accurate enough.
Parameters:
goalPoint - The relative end point (relative to Astrobee)
orientation - The absolute orientation
printRobotPosition - Flag whether to print robot positions in log or
not.
Returns:
A Result instance carrying data related to the execution. Returns null
if the command is NOT executed because of an error.

• startMission

47
 Version 1.0
Release Date: April 1st, 2025

public boolean startMission()

Undock Astrobee from docking station. Astrobee then starts counting the mission time.
Returns:
Returns True if the execution is successful. Returns false if the
command is NOT executed because of an error.

• notifyRecognitionItem

public void notifyRecognitionItem()

Notify that Astrobee has recognized the TargetItem held by the astronaut.
• reportRoundingCompletion

public boolean reportRoundingCompletion()

Report rounding completion and blink the lights.

Returns:
Returns True if the execution is successful. Returns false if the
command is NOT executed because of an error.

• takeTargetItemSnapshot

public void takeTargetItemSnapshot()

Take a snapshot of TargetItem.

When execute this method, the mission is completed.

• getNavCamIntrinsics

public double[][] getNavCamIntrinsics()

Get camera matrix and distortion coefficients of NavCam. Different values are returned on orbit and in
the simulator. *The parameter values are different between the simulator and real Astrobee in the ISS as
shown below. Therefore this API returns different value depending on whether the APK is running on
simulator or on real Astrobee.

Returns:
Array of camera parameters [camera matrix, distortion coefficients] for
NavCam. The array of the camera matrix and distortion coefficients is
as follows.

・Simulator

Camera matrix: [ 523.105750, 0.000000, 635.434258,0.000000, 534.765913,

500.335102,0.000000, 0.000000, 1.000000 ]
Distortion coefficients: [ -0.164787, 0.020375, -0.001572, -0.000369, 0.000000 ]

・Real Astrobee

Camera matrix: [ 608.8073, 0.0, 632.53684, 0.0, 607.61439, 549.08386, 0.0, 0.0, 1.0 ]

48
 Version 1.0
Release Date: April 1st, 2025

Distortion coefficients: [ -0.212191, 0.073843, -0.000918, 0.001890, 0.0 ]

• getDockCamIntrinsics

public double[][] getDockCamIntrinsics()

Get camera matrix and distortion coefficients of DockCam. Different values are returned on orbit and in
the simulator. *The parameter values are different between the simulator and real Astrobee in the ISS as
shown below. Therefore this API returns different value depending on whether the APK is running on
simulator or on real Astrobee.

Returns:
Array of camera parameters [camera matrix, distortion coefficients] for
DockCam. The array of the camera matrix and distortion coefficients is
as follows.

・Simulator

Camera matrix: [ 661.783002, 0.000000, 595.212041,0.000000, 671.508662,

489.094196,0.000000, 0.000000, 1.000000 ]
Distortion coefficients: [ -0.215168, 0.044354, 0.003615, 0.005093, 0.000000 ]

・Real Astrobee

Camera matrix: [ 753.51021, 0.0, 631.11512, 0.0, 751.3611, 508.69621, 0.0, 0.0, 1.0 ]
Distortion coefficients: [ -0.411405, 0.177240, -0.017145, 0.006421, 0.000000 ]

• saveBitmapImage

public void saveBitmapImage(Bitmap image,

java.lang.String imageName)

Save a bitmap image for debug. The maximum pixel size of an image is 1228800 (height x width) and up
to 50 images can be saved per simulation. The image is saved in Android Emulator (/sdcard/data/) as a
png file and can be download on the dashboard display.
Parameters:
image - Bitmap Images to save.
imageName - string Image name to save.

• saveMatImage

public void saveMatImage(Mat image,

java.lang.String imageName)

Save a mat image for debug. The maximum pixel size of an image is 1228800 (height x width) and up to
50 images can be saved per simulation. The image is saved in Android Emulator (/sdcard/data/) as a png
file and can be download on the dashboard display.
Parameters:

49
 Version 1.0
Release Date: April 1st, 2025

image - Mat Images to save.

imageName - string Image name to save.

• setAreaInfo

public void setAreaInfo(int areaId,

java.lang.String itemName)

Set matching of an area and an item.

Parameters:
areaId - area ID
itemName - item name

• setAreaInfo

public void setAreaInfo(int areaId,

java.lang.String itemName,
int number)

Set matching of an area, an item, and the number of an item.

Parameters:
areaId - area ID
itemName - item name
number - item number

6.2.1. Type information

Please refer the following links for information about Types implemented in
astrobee_android.
Table 6-2 Type information

Type URL

gov.nasa.arc.astrobee https://github.com/nasa/astrobee_android/blob/a8560ab0270
.Kinematis ac281d8eadeb48645f4224582985e/astrobee_api/api/src/mai
n/java/gov/nasa/arc/astrobee/Kinematics.java

gov.nasa.arc.astrobee https://github.com/nasa/astrobee_android/blob/a8560ab0270
.Result ac281d8eadeb48645f4224582985e/astrobee_api/api/src/mai
n/java/gov/nasa/arc/astrobee/Result.java

gov.nasa.arc.astrobee https://github.com/nasa/astrobee_android/blob/a8560ab0270
.types.Vec3d ac281d8eadeb48645f4224582985e/astrobee_api/api/src/mai
n/java/gov/nasa/arc/astrobee/types/Vec3d.java

gov.nasa.arc.astrobee https://github.com/nasa/astrobee_android/blob/a8560ab
.types.Quaternion 0270ac281d8eadeb48645f4224582985e/astrobee_api/a

50
 Version 1.0
Release Date: April 1st, 2025

Type URL

pi/src/main/java/gov/nasa/arc/astrobee/types/Quaternio
n.java

gov.nasa.arc.astrobee https://github.com/nasa/astrobee_android/blob/a8560ab0270
.types.Point ac281d8eadeb48645f4224582985e/astrobee_api/api/src/mai
n/java/gov/nasa/arc/astrobee/types/Point.java

7. Tips for Astrobee Characteristics

The tips for Astrobee characteristics will be provided in this section to be considered for
successful runs.

7.1. Rendering of Astrobee

Figure 8-1 indicates Astrobee is equipped with some external hardware components. The
hardware surrounded with one-foot cube (about 32 cm wide) are used in the simulator and ISS
Final Round.

Figure 7-1 External hardware components

51
Version 1.0
Release Date: April 1st, 2025

Table 7-1 External hardware components

Name Explanation
HazCam A monochrome camera for detecting obstacles within 30 cm
NavCam A monochrome camera for image data processing and taking
a photo after sending finish command
SciCam A color camera for taking a video
(The participants cannot utilize this.)
DockCam A monochrome camera for docking to the docking station
PerchCam A monochrome camera for grabbing a handrail
Flashlight Use this when reading QR code
Speaker Playing audio files. In the 4th Kibo-RPC, the speaker will not
be used.

Figures and tables below show the positions of the hardware components.

Figure 7-2 Astrobee Front View

Table 7-2 Distances from center point

x[m] y[m] z[m]
Nav Cam 0.1177 -0.0422 -0.0826
Haz Cam 0.1328 0.0362 -0.0826

52
 Version 1.0
Release Date: April 1st, 2025

Figure 7-3 Astrobee Back View

Table 7-3 Distances from center point

x[m] y[m] z[m]
Dock Cam -0.1061 -0.054 -0.0064
Perch Cam -0.1331 0.0509 -0.0166

7.2. Specification of Astrobee

⚫ Mass: 10kg (Installed only two of the four batteries)
⚫ Maximum Velocity: 0.5 m/s
⚫ Maximum Thrust (X axis): 0.6 N
⚫ Minimum moving distance: 0.05 m.
⚫ Minimum rotating angle: 7.5 degrees.
⚫ If the Astrobee detects the actual obstacles in front, it will automatically stop and then
maintain its position and orientation. The moving path it is on will also be discarded.

7.3. References of Astrobee

⚫ GitHub-1 ( https://github.com/nasa/astrobee )
⚫ GitHub-2 ( https://github.com/nasa/astrobee_android )
⚫ Website of Astrobee ( https://www.nasa.gov/astrobee )

53