Flight Guidance & Motion Planning Engineer

ABOUT THE ROLE

Harmattan AI is pushing the boundaries of high-speed autonomous flight, where guidance turns intent into a flyable path at speed. It takes a mission objective or a waypoint, produces the reference, and drives the aircraft along it. It is one of the most demanding parts of the flight software.

As a Flight Guidance & Motion Planning Engineer, working out of Lausanne, Paris, or Zurich, you will own motion planning, trajectory generation, and guidance laws, consuming imperfect, delayed inputs from the estimator and the perception pipeline and handing references to the control loop. You will weigh classical against learning-based methods, from optimization-based planning to learned and differentiable approaches, and decide what holds up at high speed.

RESPONSIBILITIES

• Motion Planning: Plan dynamically feasible paths onboard, including obstacle avoidance, replanning as the situation changes, and perception-aware planning.

• Trajectory Generation: Turn those paths into feasible references under airframe and actuator limits, handed cleanly to the control loop.

• Guidance Laws: Design the laws that steer the aircraft along the reference and hold it there.

• Interfaces & Perception Coupling: Own the input contracts for the ego-state from Navigation and the perception pipeline, and handle their latency, rate, and dropout, including behavior when an input drops.

• Mode & Failsafe Logic: Across cruise, loiter, and approach, including lost-link behavior.

• Flight Test & Diagnosis: Verify across simulation, SITL, and HITL, including Monte Carlo campaigns, then fly and iterate from logs.

• Mentorship: Depending on seniority, support the team by mentoring junior engineers.

CANDIDATE REQUIREMENTS

• Educational Background: A strong academic record with a degree in aerospace, robotics, GNC, or a related STEM field. A practical track record matters more than the specific degree.

• Motion Planning & Trajectory: Proven experience with onboard motion planning, trajectory generation, and tracking, and with guidance laws on flying platforms.

• Perception-Coupled Control: Hands-on experience closing a control loop on a perception output, handling latency and dropout.

• Learning-Based Methods: Familiarity with learning-based planning or guidance and simulation-based training, with the judgment to weigh it against classical approaches.