Our Technology

The complex interface and time-consuming nature of traditional robot programming are significant barriers for developers and automation providers. Every robot has its own programming language, and the programming interface only allows for basic functionality. This is why we have developed a state-of-the-art motion planning engine that powers the Jacobi platform. Unlike existing open-source motion planning algorithms, the Jacobi planner computes a full and dense motion plan that includes the positions, velocities, accelerations, jerk, and timing of the robot joints at every point along the robot's trajectory.

The Jacobi planner stands out by being:

• Fast. It leverages advanced Artificial Intelligence to speed up planning by 1000x, from over a second to less than a millisecond of computation time.

• Reliable. It uses complex optimization to calculate fast motions around obstacles without hitting joint limits or singularities.

• Flexible. It integrates easily with every hardware, both industrial and collaborative robots, with any reach and payload.

Based on Years of Research

Our founding team comes from robotics research at UC Berkeley, with connections to universities including Carnegie Mellon University, Imperial College London, and Karlsruhe Institute of Technology. There, we spent multiple years working on motion planning for various applications such as Ruckig for instantaneous trajectory generation.