Information flow is the best bridge we have between Relativity and Quantum Mechanics. In Relativity it's all about light cones and light clocks, in QM it's about when systems are isolated and when state information propagates from it.
I think I understand your unease. What you need is a physicists picture of continuous symmetries under transformation, which by Noether's theorem are conservation laws.
When a particle collision happens, whichever restframe you pick the total momentum is conserved in the system as a whole, above the very short timescales of the uncertainty principle. If an uncertain interaction like say a three-body interaction happens in a black box that will magnify the uncertainties (eg The three-body problem shows us why we can’t accurately calculate the past, where uncertainties below the Planck scale matter), even so we know the total momentum of the system will be maintained. Uncertainty, doesn't mean no knowledge. An interesting point is that angular momentum is also relative like this, so when it's given we implicitly rely on a general restframe for the universe, but the universe as a whole could be spinning by any amount, and it would be unverifiable, unmeasurable.
Consider a particle-antiparticle pair. We can think of space as a sea of low-excitation pairs. When energy is added to a virtual pair, like say a gamma ray, you get a real pair, and the total quantum numbers of the massless chargeless photon are maintained, by the pair. But once they move apart, there isn't uncertainty like there was for the virtual pair, charge will be conserved even though it 'appeared from nowhere'. Feynman Diagrams can help us picture the virtual particle sea, with particle interactions shaped by all the possible things that could happen, but with probabilities and so impacts on the system observables, scaled down with the number of vertexes or crossing points of particle paths in the diagram.
So, saying all the properties are relational isn't like a kind of 'total subjectivity', where anything could be anything. There are waves of information bouncing around, limited by the speed of light (except entangled information, ER = EPR, but this can't be used to transmit FTL signals, I think of it as like spinning a coin and cutting it in half while it's still spinning, like the particle pair you know what is conserved), and paused fuzzily in isolated systems - entropy and the second law of thermodynamics can be described as state information's tendency to spread out, see the purification principle.
We can picture the whole universe as arising from uncertainty alone, with net properties of charge and mass of zero, but a series of divisions and separations that have emergent symmetries, like even space and time themselves emerging from a fuzzy soup of uncertainty (the quantum spin network).
More on the limits of causal narratives vs symmetries & conservation laws here: Is the idea of a causal chain physical (or even scientific)?