Real world problems need real world solutions. We think driverless trucks will soon be driving the roads rather than stuck as a 5 year research problem. That means we prioritize practical solutions over untested theories. At Starsky, what you build will be used soon and not in a far away future
No one has done this before. Plain and simple, you will be part of the first team that solves the autonomy problem for trucks. We believe we have a good solution, but will need to adapt our approach along the way. We are always open to new algorithms and experiments as long as they are safe.
Solid implementations. We have safety mechanisms to cope with issues and processes to make sure that we are doing the right things right, but nothing can replace good engineering. The controller is a safety critical system that cycles a million times a day so the bug that happens one-in-a-million times will never meet our standard.
It's a hard problem. The truck is a multibody vehicle weighing over 40 tons and over 20m long. It's subject to poor road conditions, wind, wake of other vehicles, different load configurations and more. Being off the path by one tire width means we are already touching the lane boundaries. We are looking for people who are not afraid of challenges.
Not sure if you have the skillset?
These are the first things we look for on resumes:
A strong background in vehicle dynamics and control theory
Professional experience in linear and non-linear control techniques such as LQR/LQE/LQG, MPC, Adaptive Control or Robust Control
PhD or >5 years working with Controls in safety-critical, embedded or real-time systems.
These are the first things that are not needed, but we like seeing on resumes:
Experience with multibody systems
Experience with numerical optimization techniques
Familiarity with python
Experience programming in C++
Starsky Robotics
