Physical AI for Space Logistics and Transportation
Space logistics is a velocity budget you compose and a chain of maneuvers an autonomy stack flies. Compute the rocket-equation wall and pay it down with infrastructure; propagate relative motion with Clohessy-Wiltshire and solve a two-impulse rendezvous; estimate a tumbling target with a Kalman filter; and seat a part by force, not position. Every bench runs the real math behind the Charlot Lab's orbital-logistics instruments.
▶ Start the course ← All coursesWhere this sits, and what moves it.
Binding constraint · Delta-v and the energy to produce it. In space every decision is denominated in a budget that does not refill, and autonomy is what you buy when a round trip to a human costs more than the manoeuvre.
Logistics off-planet meant a mission designed years in advance around a fixed manifest. Re-planning against something unexpected was a ground operation with a light-time penalty attached.
On-board planning against a real delta-v budget is buildable, and this course builds it. What is not solved is verification: a policy that re-plans is a policy whose behaviour was not reviewed before launch, and no agreed method exists for certifying one -- which is the same evidence gap that appears in every unattended system.
What changes it is a machine that can demonstrate to someone who was not there what it did and why, cheaply enough to do it for every manoeuvre. Watch for evidence formats rather than autonomy demonstrations; autonomy is arriving on its own, and the ability to check it is not.
Every hard thing was impossible until the constraint that made it impossible was named. How we read a frontier →
The Velocity Budget
Turn every destination into a delta-v budget, see the rocket-equation wall, and pay it down by composing shared infrastructure with the onboard vehicle.
- L2The rocket-equation wallCan a single launch vehicle carry a useful payload to the surface of Mars on its own?See why a single vehicle paying the whole velocity budget hits an exponential wall, and where a destination becomes physically unreachable.→
- L2Composing the budgetYour payload fraction comes out negative. What does that mean physically?Pay the budget down with reusable infrastructure and autonomous in-space transport, and reach destinations a single vehicle cannot.→
Rendezvous & Proximity Operations
Reason in the target's local frame: propagate relative motion with Clohessy-Wiltshire and solve the two-impulse maneuver that flies a chaser to a soft berth.
- L3Relative motion: Clohessy-WiltshireA chaser sits near a target on a circular orbit with a small along-track offset and no relative velocity. What happens?Propagate a chaser's motion near a target on a circular orbit, and discover the two behaviours that make rendezvous non-obvious.→
- L3Two-impulse rendezvousYou solve for a two-impulse transfer and shorten the flight time. What happens to the total delta-v?Solve for the burn that flies a chaser to the target in a chosen time, price both burns, and find the faster-costs-more tradeoff.→
Capture, Servicing & Assembly
Close the last links with autonomy: estimate a non-cooperative target with a Kalman filter, and seat a part by force where precision fails.
- L3Estimating a non-cooperative targetYour Kalman filter tracks a tumbling target worse than using the raw measurements. Where is the bug?Track a noisy, tumbling target with a constant-velocity Kalman filter and show it beats using the raw measurements.→
- L4Seat it by force: compliant servicingYou insert a part into a socket with lateral misalignment, under rigid position control. What happens?Insert a part into a socket with misalignment, and show that force control seats it where rigid position control jams.→