Multi-robot systems require the embedding of attributes such as cooperation, coordination, negotiation, leadership, autonomy, smartness, and cognition in their systems design that extends to both their hardware and software. A number of undergraduate students were involved in the design and implementation of the first generation of resilient multi-robot systems over a period of 5 years as part of either funded research projects or final-year dissertation project work. This paper presents results highlighting their potential efficacy in significantly enhancing their systems design, integration and programming skills through instructor-mediated active learning using teams of physical and simulated robots and drones. A number of educational robots and drones such as Turtlebots, Diddyborg, CoDrone Pro, DJI Tello EDU, and simulation environments in Ubuntu, Raspbian, Windows, etc. such as Gazebo and ConstructSim were used and their cognitive pedagogical effects related to perception, imagination, memory, and problem-solving during and after these student projects are discussed. The case studies from these projects show that they enhanced the communication skills, insight, and reasoning of the students leading them to innovate within their projects and simultaneously deliver on industry-relevant outcomes.