Is anyone working on zk-based consensus? Intuitively it seems like you could replace zk-unfriendly signatures in consensus algorithms with (recursive) proofs of pubkey = hash(privatekey, data) using a zk-friendly hash, which would enable much cheaper / faster proofs.

Mina is the closest I've seen to this. Doesn't it use a curve-based signature scheme for accounts though?

I'm wondering if replacing curve-based signature verification with hash-based proof verification might result in simpler / cheaper / faster proofs.