GitHub  •  LinkedIn  •  X  •  Discord

Telekinesis Synapse

Telekinesis Synapse is a Python SDK for industrial robot control, providing a unified API across leading robot brands with native support for real hardware and NVIDIA Isaac Sim.

Key features:

• Unified Skills across all supported brands
• Offline kinematics: FK and IK work without any hardware or simulator connected
• Live robot state publisher over BabyROS for real-time state streaming

Table of Contents

• Telekinesis Synapse
  • Table of Contents
  • Installation
  • Getting Started
  • Example Usage with Real Hardware
  • Resources
  • Support
  • License

Installation

Requirements: Python 3.12, Windows / macOS / Linux.

1. Create and activate a Conda environment:

   conda create -n telekinesis-synapse python=3.12
   conda activate telekinesis-synapse
   

2. Install telekinesis-urdfs (kinematic data):

   cd /path/to/working/directory
   git clone --depth 1 https://github.com/telekinesis-ai/telekinesis-urdfs.git
   cd telekinesis-urdfs
   pip install .
   

   Note: telekinesis-urdfs is a large repository. The initial clone and wheel build are expected to take several minutes. Do not interrupt the process.

3. Install Telekinesis Synapse:

   pip install telekinesis-synapse
   

Getting Started

Create a robot and read its state offline (no hardware or simulator needed):

from telekinesis.synapse.robots.manipulators import universal_robots

robot = universal_robots.UniversalRobotsUR10E()

print("Joint positions (deg):", robot.get_joint_positions())
print("TCP pose (m, deg):    ", robot.get_cartesian_pose())

Move the robot to a target joint configuration offline:

target = robot.default_joint_configuration.copy()
target[0] += 15.0  # rotate base joint 15°
robot.set_joint_positions(target)
print("New TCP pose:", robot.get_cartesian_pose())

Example Usage with Real Hardware

Prerequisites:

1. Robot powered on and connected to the same network as your machine.
2. PC on the same subnet as the robot (e.g. robot at 192.168.1.100, PC in 192.168.1.0/24). Verify with ping 192.168.1.100.
3. Remote control enabled on the teach pendant.

from telekinesis.synapse.robots.manipulators import universal_robots

robot = universal_robots.UniversalRobotsUR10E()
robot.connect(ip="192.168.1.100")

print("Joint positions (deg):", robot.get_joint_positions())
print("TCP pose (m, deg):    ", robot.get_cartesian_pose())

current_tcp_pose = robot.get_cartesian_pose()
target_tcp_pose = current_tcp_pose.copy()
target_tcp_pose[2] += 0.05 # move 1cm in z
robot.set_cartesian_pose(target_tcp_pose)

robot.disconnect()

For more examples and full documentation see the Resources section.

Resources

• Telekinesis Examples: runnable examples for every supported robot and tool.
• Documentation: full API reference and guides.

Support

• Open an issue on GitHub.
• Email: support@telekinesis.ai
• Discord

License

Proprietary. © Telekinesis. All rights reserved.