DAS Ego Manual v1.0

简体中文 | English

1. Important Safety Information

1.1 Safety Instructions

Please read this section carefully and follow all safety guidelines before first use.
For assistance, contact: support@genrobot.ai

1.1.1 Warning Symbol Meanings

• ⚠️ Danger: May cause serious injury, death, or major property damage.
• ⚠️ Warning: May cause personal injury or equipment damage.
• ⚠️ Caution: May cause equipment failure or data loss.
• 💡 Tip: Helpful suggestions for better operation.

1.2 Liability and Limitations

• Do not modify, alter, or tamper with the device.
• GENROBOT.AI is not responsible for issues caused by misuse, unauthorized modifications, or improper operation.
• By using this device, you agree to all safety terms and assume full responsibility for its operation.
• This product is not intended for users under 18 years old.

1.3 Integrator and User Responsibilities

• Perform a complete hazard and risk assessment before deployment.
• Implement appropriate safety protections based on the assessment results.
• Ensure all hardware and software are installed and configured correctly.
• Do not disable or change safety measures without authorization.
• Comply with applicable laws, standards, and industry regulations.

1.4 Environmental Requirements

• Operating temperature: 0°C to 40°C (optimal 22±2°C)
• Relative humidity: 40%RH to 65%RH (non-condensing)
• Protection rating: IP50
• Magnetic field strength: ≤50 mT
• Illumination: ≤500 lux, with no direct light or reflective surfaces
• The ground should be vibration-free and away from motors, frequency converters, high-power supplies, and WiFi 6E/5G base stations
• The environment should be dust-free, smoke-free, and free of corrosive gases

1.5 Usage Warnings

1.5.1 Danger Warnings

• Do not modify or tamper with the device.
• Stop using the device immediately if it is damaged or unstable.
• Avoid exposing the device to strong magnetic fields for extended periods.

1.5.2 Operational Warnings

• Implement safety measures according to the risk assessment.
• Ensure bystanders stay away from the device's operating range.
• Reassess risks when integrating new components or equipment.
• Read all related equipment manuals before operation.

1.5.3 Maintenance Warnings

• Use only original accessories and replacement parts.
• Inspect the device regularly to identify potential issues early.
• Follow the recommended maintenance schedule.
• Store the device under compliant environmental conditions.

1.6 Data Security and Privacy

1.6.1 Data Protection

• Handle collected data securely, as it may contain sensitive information.
• Back up important data regularly.
• Understand the data policies of any connected cloud services.
• Comply with applicable data protection regulations.

1.6.2 Privacy Protection

• Respect the privacy of others when operating in public spaces.
• Obtain necessary permissions before recording or collecting data.
• Follow local laws regarding data collection.

2. Product Overview

2.1 Key Features

DAS (Data Acquisition System) is a general-purpose embodied AI multimodal data acquisition system.
DAS Ego is a first-person, head-mounted multimodal data acquisition device in the DAS product family and serves as an entry point for ego-centric data collection. It integrates six high-resolution cameras, a high-precision IMU, microphone audio, and multiple sensors to provide ultra-wide visual coverage and millimeter-level trajectory reconstruction for high-quality data capture. It also supports multi-device spatiotemporal alignment and voice-controlled recording, making it suitable for long-duration collection tasks.

2.2 Packing List and Specifications

Item	Quantity	Specification
Ego main unit	1	269 g
Battery pack	2	4000 mAh, 3.7 V, 100 g each
TF card	1	128 GB
Card reader	1	USB 3.0, dual-slot
Charging cable	1	USB to Type-C, 2 m

2.3 Device Layout

Module	Layout / Specification
Cameras	Six wide-angle RGB cameras with approximately 2 MP each; horizontal FOV about 270°, vertical FOV about 150°; recording at 30 Hz.
Battery	Magnetically attached to the DAS Ego main unit; 4000 mAh per pack, about 2 hours of runtime; batteries can be swapped without powering off.
IMU	High-precision 6-axis sensor with a 2×2 IMU array; after calibration and algorithm processing, gyro bias stability is about 2.5°/h, and drift over −40°C to 85°C is about 0.03°/s.
Coordinate system	The device kinematics and coordinate definitions follow the Feishu URDF document.

2.4 Device Overview

2.5 DAS App Introduction

This app is the dedicated companion application for the Ego head-mounted data acquisition device. It supports both Android and iOS. The app is the core terminal for data capture, device management, and data upload, enabling efficient and accurate data collection with a convenient workflow suitable for a variety of collection scenarios.

2.5.1 DAS App Installation Guide

For Android:

1. Save the APK package to your phone. The latest version can be found in the version update document.
2. Install the package.

For iOS:
Search for genrobot in the App Store, then download and install it.

2.5.2 DAS App Core Functions

The app's core functions include device connection, device management, data upload, and other utilities.

• Device connection: through the four steps of Device Management, Add Device, serial number confirmation, and WiFi configuration, you can quickly complete Bluetooth, WiFi, and SD card setup to ensure stable interaction between the device and the app.
  • Turn on Bluetooth.
  • Turn on WiFi or a mobile hotspot.
  • Power on the Ego device.
  • Open the app-Genrobot.
  • Connect the device: tap Device Management -> Add Device -> Connect for the target device -> Configure WiFi.

Device Management

Add Device

Connect Device

Configure WiFi 1

Configure WiFi 2

• Device management
  • After the device is connected, you can view its Bluetooth status and network address in the app.
  • Device information: product model, firmware version, CPU number.
  • Device status: Wi-Fi, SD card, SD card capacity, recorded item count, pending upload count, recording duration, and battery level.
  • Firmware upgrade.

Bluetooth and IP

Device Info and Status

Firmware Upgrade

• Data upload
  App upload is not supported in the user version.

• Other

  • Advanced function - SD card formatting: after recorded data has been uploaded, or whenever you want to clear the SD card, you can format it.
  • Delete device or restart device.
  • Reconfigure the network.
  • Settings: choose theme color and language.

SD Card Formatting

Restart/Delete Device

Reconfigure Network

Settings

Settings Detail

3. Data Collection Guide

3.1 Wearing the Device

1. All hair must be tied up with a hairband or clip to avoid blocking any camera lens.
2. Clean the lenses to keep the image clear.
3. Wear the device correctly: keep it level and close to the eyebrows, with the logo centered on the forehead.

Proper Wearing Guide

3.2 Powering On and Off the Ego Device

• Press the power button once. When you hear Power on, the device is powered on, and the power indicator turns green.
• Press and hold the power button. When you hear Power off, the device is powered off.

Power Button

Power Indicator and SD Card Slot

3.3 Recording Data

• Press the power button once. When you hear Start recording, recording has started.
• Press the power button again. When you hear End recording, the current recording ends.

3.4 Other Notes

• When the battery is low, you will hear the voice prompt Low battery, indicating that the battery should be replaced.
• If no SD card is inserted, recording cannot start, and the voice prompt No sdcard will be played.

4. Data Processing Guide

1. After DAS Ego data is recorded, the raw data is saved as mcap files on the SD card. For details about the mcap format, see the official site: https://mcap.dev/. For the complete mcap proto format, see https://github.com/genrobot-ai/das-datakit/tree/main/proto.

2. After recording is complete, remove the SD card and use the included card reader to copy the data to your computer.

3. Use the visualization tool https://monitor.genrobot.click/#/ to open mcap files for visualization.

• Open local data.
• After opening the data, Ego data will be selected automatically.

Open Local Data

Auto-select Ego Data

• For an introduction to the main interface, see Monitor Introduction.
• Open the details page to view the recording results.

Detail View

Topic List (Raw Data)

4. Post-processing

• Image download

  Run the following commands to pull the image:

  docker pull 镜像仓库域名/<命名空间名称>/<OCI 制品仓库名称>:[镜像版本号]
  docker pull imagepublic.genrobotai.com/genrobot/ego_slam:v0.1.21

• Image usage

  1. Create a working directory.

  2. Download init.sh and run.sh from: https://zcnma1sv5kma.feishu.cn/docx/ZRlodJBN6ortulxKW5XczXgfnch

  3. Place init.sh and run.sh in the working directory as shown below:

     

     Reference Working Directory

  4. Create a data folder inside the working directory.

  5. Copy the target mcap data into the data directory.

  6. Run init.sh.

  7. Run run.sh. A successful run looks like this:

     

     Successful Run Example

  8. Enter the container and run the script:

     Script 1: process all data in the folder and generate results with ground-truth pose.

     bash /app/scripts/pipeline.sh --device-type ego --input-dir /app/data/ --output-dir /app/data/output/

     Script 2: process all data in the folder, generate results with ground-truth pose, run additional checks, and return check results.

     bash /app/scripts/pipeline_checker.sh --device-type ego --input-dir /app/data/ --output-dir /app/data/output/

  9. Verify the output: the vio trajectory will be written into the generated mcap file, and checker will validate the generated mcap directly, as shown below:

     

     Output Verification

     Error Test:

     errorcode(Status)	Check item	Parameters	Description	Priority	Notes
     00000000	Overall success	-Success	All enabled checks passed	-	Enabled
     00000023	Trajectory length	-t, --tracker-threshold 10cm	Trajectory length did not exceed the threshold	P0	Enabled
     00000024	Trajectory coordinate range	-e, --eef-pose-threshold	Trajectory coordinate range exceeds the threshold	-	Not enabled
     00000025	Trajectory frame drops	-s, frame-drop threshold settings	Frame drops exceed the threshold	P0	Enabled
     00000027	Trajectory smoothness	-j, --jerk	Smoothness check failed	P0	Enabled
     00000031	Input file validity	File path / format / content	Input file error	P1	Enabled
     00000032	Configuration validity	Check switch configuration	No checks enabled	P1	Enabled
     00000033	Runtime exception	Internal processing logic	Processing error	P1	Enabled

  10. Visualize the result at https://monitor.genrobot.click/#/index. Load the generated mcap data in the visualization page to inspect the output.

5. Coordinate Systems & Topic Introduction

5.1 Topic Introduction

Main topics:

Topic	Description	Frequency
/robot0/sensor/camera0/compressed	cam0 image stream, H.264 encoded	30 Hz
/robot0/sensor/camera1/compressed	cam1 image stream, H.264 encoded	30 Hz
/robot0/sensor/camera2/compressed	cam2 image stream, H.264 encoded	30 Hz
/robot0/sensor/camera3/compressed	cam3 image stream, H.264 encoded	30 Hz
/robot0/sensor/camera4/compressed	cam4 image stream, H.264 encoded	30 Hz
/robot0/sensor/camera5/compressed	cam5 image stream, H.264 encoded	30 Hz
/robot0/sensor/imu	IMU information	200 Hz
/robot0/vio/eef_pose	VIO end-effector pose	30 Hz
/robot0/vio/relative_eef_pose	Relative VIO end-effector pose	30 Hz
/robot0/sim/robot_info	URDF robot info: end-effector pose and joints	30 Hz
/robot0/sensor/camera0/camera_info	Calibration info	1 frame
/robot0/sensor/camera1/camera_info	Calibration info	1 frame
/robot0/sensor/camera2/camera_info	Calibration info	1 frame
/robot0/sensor/camera3/camera_info	Calibration info	1 frame
/robot0/sensor/camera4/camera_info	Calibration info	1 frame
/robot0/sensor/camera5/camera_info	Calibration info	1 frame

Notes:

• camera0 to camera5 are arranged from left to right in the Ego top view.
• /robot0/vio/eef_pose records the DAS Ego pose in the world frame (Tilted state in the figure below).
• /robot0/vio/relative_eef_pose uses the DAS Ego pose at time t=0 as the world-frame origin and records subsequent pose changes relative to that initial state (Aligned state in the figure below).

5.2 World Frame

• Origin: IMU position at the start of capture.
• Z axis (blue): Opposite to gravity, pointing upward.
• X axis (red): Forward projection of the IMU orientation onto the horizontal plane after successful initialization.
• Y axis (green): Orthogonal to X and Z, defined by the right-hand rule.

World Frame

5.3 Body / Local Frame

• Origin: IMU geometric center.
• X axis (red): Device forward, along the normal of the IMU mounting plane pointing forward.
• Y axis (green): Device left, orthogonal to X, per the right-handed convention.
• Z axis (blue): Device up, orthogonal to both X and Y, uniquely defined by the right-hand rule.

Local Frame