Optimal Decision-Making for Autonomous Agents via Data Composition

We consider the problem of designing agents able to compute optimal decisions by composing data from multiple sources tackle tasks involving: (i) tracking a desired behavior while minimizing an agent-specific cost; (ii) satisfying safety constraints. After formulating control problem, we show that this is convex under suitable assumption and find solution. The effectiveness results, which are turned in algorithm, illustrated on connected cars application via in-silico in-vivo experiments with real vehicles drivers. All confirm our theoretical predictions deployment algorithm vehicle shows its suitability for in-car operation.