This paper presents algorithms for performingdata-driven reachability analysis under temporal logic sideinformation. In certain scenarios, the data-driven reachablesets of a robot can be prohibitively conservative due to theinherent noise in the robot’s historical measurement data. Inthe same scenarios, we often have side information about therobot’s expected motion (e.g., limits on how much a robotcan move in a one-time step) that could be useful for furtherspecifying the reachability analysis. In this work, we showthat if we can model this side information using a signaltemporal logic (STL) fragment, we can constrain the datadriven reachability analysis and safely limit the conservatismof the computed reachable sets. Moreover, we provide formalguarantees that, even after incorporating side information, thecomputed reachable sets still properly over-approximate therobot’s future states. Lastly, we empirically validate the practicality of the over-approximation by computing constrained,data-driven reachable sets for the Small-Vehicles-for-Autonomy(SVEA) hardware platform in two driving scenarios.