SDK & teleoperation

Get comfortable with GentoPlatform first; understanding how the robot behaves makes code development easier.

Gento SDK

The motion control SDK wraps the low-level communication (L0) and provides interfaces for system management, motion control, state switching, parameters, kinematics, and trajectory planning.

Resource	Link
Gento series SDK repository	tianjizn/tianji-robot-SDK (Gento_Skye+Luna branch)
Motion control SDK API docs	tianjizn.github.io/api
Documentation source repository	tianjizn/tianjizn.github.io

If the API docs are not reachable online, download them from the documentation repository and view them locally.

Directory structure

GENTO_SDK/
├── C_SDK/                  # SDK source (Common, FileClient, Kinematics, L0Control, L1Robot)
├── C_EXAMPLE/              # C++ examples that call C_SDK source directly
├── C_EXAMPLE_USE_DLL_SO/   # C++ examples that call the compiled DLL/SO
├── PYTHON_SDK/             # Python wrapper (GentoRobot.py)
├── PYTHON_EXAMPLE/         # Python examples
├── win_auto_compile.bat    # One-step Windows build (source → DLL)
├── linux_auto_compile.sh   # One-step Linux build (source → SO)
└── README.md

Version compatibility

Controller and SDK versions are MAJOR.MINOR.PATCH. The SDK connects only when MAJOR and MINOR match; otherwise the connection returns error -4 "Version incompatible". Call FX_L1_System_GetSDKVersion() (Python: get_sdk_version()) to get the SDK version, then upgrade or downgrade the controller to match.

Core modules

Module	Description
System management	Connect/disconnect, log level, reboot, firmware update, file transfer
State machine	Switch between position, impedance (joint / Cartesian / force), drag teaching, and collaborative release
Real-time feedback	1 kHz data (joint position, velocity, torque, IMU, F/T) and 500 Hz slow-group data
Parameters	Read and write parameters by name
Terminal communication	Send and receive with external devices over CAN FD or RS485
Hardware configuration	Brake lock/release, encoder offset reset, soft-limit disable
Runtime motion	E-stop, joint position commands, force/torque control, stiffness/damping
Kinematics & planning	FK/IK, Jacobian, tool transforms, MoveJ/MoveL, multi-segment Cartesian, dual-arm sync
Dynamics identification	Identify payload mass, center of mass, and inertia from recorded data

Usage notes

• SDK demo logic and parameters are for R&D reference only, not production code.
• Stiffness and damping values are references and may change between controller versions; ask Gento support.
• The SDK uses reliable UDP. Allow UDP on the default ports 50000–50010 in your firewall.
• FX_L1_System_Link() returns a positive latency in microseconds on success; negative values are errors.
• Always call FX_L1_System_Unlink() before your program exits.

Development environment

• Debug computer: Windows or Ubuntu 20.04 x86, IP in the 6.6.7.x subnet
• Python 3.10 or later; build contrlSDK and kinematicsSDK and put libMarvinSDK.so and libKine.so in the matching SDK_PYTHON directories

Check connectivity:

ping 6.6.7.6    # chassis
ping 6.6.7.190  # motion controller
ping 6.6.7.100  # domain controller

Workflow: read the API docs → follow C_EXAMPLE / PYTHON_EXAMPLE → test at low speed and small range → add complexity step by step → integrate.

For chassis-level development (velocity control, status, sensor data, navigation and obstacle avoidance), see the general robot software interface document supplied by Tianji.

VR teleoperation

The teleoperation system lets an operator control the robot immersively with a VR headset and controllers. It also includes a toolchain (KM Data Converter) for turning recordings into imitation-learning datasets.

Part	Description
Front end	Host application for Ubuntu and Windows; preinstalled on the domain controller
Back end	Runs on the Orin domain controller: ROS2 nodes, motion control, cameras, and recording
VR headset	Meta Quest and Pico; the teleoperation app is preinstalled

Back-end services: apex-backend.service (HTTP/WebSocket entry), apex-camera.service (four GMSL cameras), apex-robot.service (ROS control and state), apex-teleop.service (IK, planning, command multiplexing).

Packages and documentation:

• Server: KernelMind_Apex_Deb
• Front end: KernelMind_Apex_Web_Deb
• Headset APK: KernelMind_Apex_VR_Apk
• Apex documentation center

Teleoperation steps

1. After every Orin / Thor power-up, initialize the cameras: log in to the domain controller remotely, run cd ~/cam_geac && ./rb_camera.sh, and wait about 20 seconds.
2. Open the teleoperation front end, enter the controller IP, and connect. Start dnsmasq (for a wired headset) and Robot, and check that the URDF pose matches the real robot. Then start Teleop, and Camera and Tool as needed.
3. Click Start Robot to reach Ready, then Impedance Mode, then Home to go to the teleoperation start pose. With a dexterous hand, switch Input Mode to Teleop.
4. Put on and calibrate the body trackers. In the headset client, confirm height, IP, and end effector, then connect. When the front end's VR light turns green, press Y on the left controller to start, and hold the side grip to operate.

Teleoperation rules

• Use a wired network; teleoperating over headset WiFi is not recommended.
• Before starting, check the E-stop, controller power, network cables, USB drive, and headset battery; keep people clear and the E-stop within reach.
• Watch the real robot, not just the 3D model, when switching modes or running Home.
• Switch to standby when not teleoperating. Back up configuration files before changing IP, payload, stiffness, damping, Home, or TCP offset.
• Stop immediately if the headset runs out of battery, loses power, or shuts down.

Recording, playback, and logs are covered in the Apex documentation center. Before replaying on the real robot, clear the workspace and confirm the start pose matches the recording.

Next: maintenance & safety.