Presenting at AVEC'26 in Tsukuba

Next week I'll be in Tsukuba, Japan, for the 17th International Symposium on Advanced Vehicle Control — AVEC '26, running from 7 to 11 September at the Tsukuba International Congress Center.

I'm presenting Linear Quadratic Regulator Control for Integrated Thermal Energy Management in Battery Electric Vehicles.

Why this problem

Thermal management is one of the quieter levers on electric vehicle range. The battery wants a narrow temperature band, the cabin wants comfort immediately, and the power electronics want to stay under their limits — and every one of those demands draws from the same pack.

Model predictive control handles that coupling well, and most of my research so far has gone that way. But it asks a lot of the hardware, and the question that keeps coming back is not whether the optimum exists — it's how much you give up by making it computable on a production ECU.

The part I find interesting

This isn't a textbook LQR with weights picked by hand until the response looks reasonable. The weights come from the Hessian of the power curve, so the quadratic cost is a local second-order approximation of the actual energy consumption rather than a tuning proxy for it.

That's the trade I care about: the controller stays cheap enough to evaluate at every step on real hardware, but the thing it optimizes is tied to the physics of what the system actually costs to run — instead of a set of gains that happened to work on the test cases.

If you'll be there

AVEC has been running since 1992 and is one of the better venues for vehicle dynamics and control work, with a genuine mix of industry and academia. If you're attending, I'd be glad to talk — about thermal management, predictive energy optimization, or anything adjacent.

The AVEC proceedings are published by JSAE with open access on J-STAGE, so the paper will be publicly readable.