A Robotic Summer: Pascal Completes UMD’s Rigorous SPARC Program
A rigorous summer at UMD SPARC moved Pascal beyond introductory robotics into the demanding work of building, programming, testing, and presenting an autonomous machine.
Congratulations to my son, Pascal, on completing the University of Maryland’s Maryland Robotics Center: Student Preparation to Advance Robotics in College, or SPARC, Certificate Program. After three intensive weeks of building, wiring, programming, testing, debugging, and presenting an autonomous ground vehicle, he received his certificate during the final showcase at A. James Clark Hall on July 24.
The photograph captures a proud moment, but the certificate represents much more than attendance at a summer camp. SPARC placed high school students inside a university engineering environment and introduced them to work resembling the early stages of collegiate robotics education. Students assembled physical systems, integrated electronics, calibrated sensors, developed control logic, processed information, and tested whether their machines could navigate without human intervention. The program ran from July 6 through July 24 at the University of Maryland, College Park, with full instructional days devoted to laboratory work, demonstrations, technical development, and team projects.
Calling SPARC a "robotics camp" is technically accurate, but the phrase creates the wrong impression. The program was not simply a recreational introduction to robotics, nor was it an elaborate version of LEGO robotics. LEGO systems have genuine educational value. Their modular parts and standardized connections help younger students learn mechanisms, programming, teamwork, and logical problem solving. SPARC began where that introductory model leaves off.
Students confronted the more difficult task of integrating mechanics, electronics, sensors, motors, motor controllers, power systems, programming, perception, planning, localization, and feedback control. Components did not simply snap into a closed educational ecosystem. Each robot had to function as a physical and computational system whose parts could fail separately or together. A loose connection could interrupt a signal. A sensor could require calibration. Motors could respond differently to identical commands. Code could work in simulation yet fail when transferred to a physical machine. Students had to determine whether a problem came from hardware, software, sensing, mechanical design, or an interaction among them. That responsibility marks the difference between an introductory robotics activity and authentic engineering education.
From Assembly to Autonomous Navigation
The SPARC curriculum followed the sequence by which an engineered system gradually acquires capability. During the first week, students formed teams, toured Maryland Robotics Center laboratories, discussed mobile robot designs, developed build concepts, created models for 3D printing, and assembled the robot body. They then studied wiring diagrams, mounted motors and electronics, integrated motor controllers and microelectronics, installed sensors and power systems, tested the hardware, verified the design, and calibrated the machine.
The order mattered. Students could not treat software as a magical layer placed over an unknown device. They first had to understand the robot as a physical construct. Motors convert electrical energy into motion. Motor drivers regulate that motion. Sensors provide limited information about the environment. Power must reach each component reliably. Mechanical geometry affects turning and distance. Calibration connects numerical commands to physical behavior.
A central engineering lesson emerged early: intelligence depends on infrastructure. A robot cannot plan effectively if its sensors produce unreliable information. It cannot execute a plan if its motors respond unpredictably. It cannot estimate its own movement if its physical structure introduces errors that the software does not recognize.
During the second week, students moved from construction into control and perception. They studied the programming codebase, developed basic motion commands, compared open loop and feedback control, tuned the robot’s behavior, processed sensor data, estimated pose, detected walls, followed lines, centered the robot, and avoided collisions. A maze mini challenge then required teams to assess their results and redesign elements that did not perform as intended.
Open loop and feedback control illustrate the program’s rigor. An open loop system issues a command without checking whether the intended result occurred. A robot may receive instructions to drive forward for a fixed period under the assumption that the command will produce a predictable distance. Reality often violates that assumption because battery levels change, motors differ, wheels slip, and surfaces vary.
Feedback control introduces measurement and correction. The system compares the intended result with the observed result and adjusts its behavior. Implementation requires students to reason about error, stability, response, and tuning. Those ideas reach beyond programming syntax. They ask students to understand a dynamic system operating in the physical world.
Perception creates a related challenge. A sensor does not provide knowledge. It provides a signal that software must interpret. A distance sensor may indicate that a wall is nearby, but the robot must determine how nearby, at what angle, and with what degree of confidence. Readings may conflict, environmental conditions may affect performance, and sensor noise may suggest changes that did not actually occur. Autonomous behavior depends on transforming imperfect measurements into useful decisions.
During the third week, the course advanced into autonomy. Students studied environmental representations, robot states and configurations, planning, simulation, localization, artificial intelligence tools, system integration, autonomy testing, tuning, and performance improvement. They also prepared final technical presentations and demonstrations.
SPARC organized autonomy around three central capabilities: perception, planning, and control. Perception allows the robot to interpret sensor data. Planning allows it to choose an action. Control allows it to execute that action accurately. Localization connects all three because a robot must estimate where it is before deciding where to go. Wheels slip, turns fall short, and sensor readings fluctuate. A successful machine must continually compare its expected position with new evidence and revise its estimate.
The final design challenge, called the Great Maze Escape, required students to integrate the entire curriculum. Working in teams, they designed, assembled, and programmed an autonomous ground vehicle that had to navigate a maze without human assistance. Each robot had to sense the environment, reason about its position and state, plan its movement, control its motors, and reach the goal. Teams received three experimental runs through a previously unseen maze, with performance measured through completion time and the number of collisions.
The maze created an elegant final examination because its objective appeared simple: find the exit. Simplicity of purpose did not reduce the complexity of execution. Strong code could not compensate for poorly calibrated motors. A mechanically sound robot could still fail if its perception system misread a wall. Accurate sensors could not rescue a planner that selected ineffective actions.
No student could reach into the maze and correct the robot during the run. Human intelligence had to be embedded in the machine beforehand through design, programming, testing, and refinement. A remotely controlled machine extends the actions of a human operator. An autonomous machine must interpret conditions and select actions for itself within the limits established by its designers.
SPARC’s alternative track showed that the program’s rigor extended well beyond autonomous vehicles. The second option focused on 3D nanoprinting and additive manufacturing for soft medical robotics. Students worked at the intersection of advanced fabrication, biomedical engineering, materials science, and robotics. Rather than producing ordinary desktop 3D printed objects, they used specialized methods connected to university research and presented how their designs could serve medical purposes.
Families saw the results during the final showcase. Students explained what they had fabricated and how nanoprinted soft robotic structures could support medical applications. Their presentation demonstrated an essential feature of engineering education: technical sophistication gains purpose when students connect it to a practical human need. Families also learned that work associated with the track had contributed to a paper accepted for presentation at a scholarly conference, an impressive measure of the program’s ambition.
The two tracks approached robotics from different directions. The Autonomous Ground Vehicles track asked how a machine can perceive, plan, and move through an uncertain environment. The nanoprinting track asked how engineers can fabricate extremely small and complex robotic structures for medical use. Both required students to move beyond demonstrations and explain how engineering decisions could solve real problems. Both treated them as emerging engineers rather than passive observers.
The final presentations reinforced that expectation. Teams had to explain their work, not merely demonstrate it. The curriculum included technical presentation preparation before the final project demonstrations and certificate ceremony. Engineering depends on such communication because a result gains value when others can understand the method, evaluate the evidence, and reproduce or improve the design.
Pascal Yglesias receives his certificate after completing the final team project of the Autonomous Ground Vehicles track of the University of Maryland’s Maryland Robotics Center SPARC Certificate Program, July 24, 2026.
Why Bridge Programs Matter for Physics and Engineering
SPARC illustrates why bridge programs matter for the future of STEM education. A bridge program connects two educational worlds. High school students usually learn established concepts through textbooks, assignments, examinations, and structured laboratory exercises. University science and engineering introduce problems that remain open, equipment that behaves imperfectly, results that require interpretation, and projects that may demand days or weeks of testing.
Strong grades in mathematics and science do not automatically prepare someone to diagnose a physical system, manage an extended technical project, collaborate under pressure, or explain an unexpected result. Students develop those abilities through practice, and bridge programs provide that practice before college begins.
Experimental physics especially benefits from this form of preparation. Physics often appears first through equations describing ideal systems. Experimental work begins when those equations encounter instruments, measurement, noise, calibration, uncertainty, and error. A theoretical prediction may be exact. An instrument never is.
Researchers must determine whether an unexpected result reveals new physics or a problem with the apparatus. They align components, establish controls, measure uncertainty, isolate variables, and repeat trials. Robotics cultivates the same discipline. When a robot turns incorrectly, students must ask whether the code contains an error, a motor responds inconsistently, a sensor is misaligned, the power supply has weakened, or several small imperfections have combined to create the failure. Guessing is not enough. Students must form hypotheses, test them, and follow the evidence.
SPARC’s syllabus identified iterative engineering development as central to the experience. Teams designed, tested, tuned, adjusted, and redesigned their robots. The final challenge required them to debug complex systems under pressure and treat obstacles as problems waiting to be solved. Failure therefore acquired a different meaning. In many academic settings, failure appears as a verdict. Engineering transforms failure into information. A collision reveals something about perception or control. Unstable motion exposes a need for better tuning. Engineers do not celebrate malfunction. They study it because failure identifies where the design must improve.
Bridge programs also help students discover whether they enjoy the actual work of STEM. Robotics looks exciting from a distance, but building an autonomous robot involves long periods of wiring, calibration, debugging, revision, and repeated testing. Experimental physics demands similar patience. Much of scientific work occurs before the dramatic result, inside the careful labor required to make an instrument trustworthy. Interest becomes more credible when it survives the difficult work behind the attractive result.
University settings deepen the experience. SPARC students worked inside the Maryland Robotics Center with faculty, professional staff, laboratories, equipment, and technical expectations. Professor Yancy Diaz Mercado, Associate Professor of Mechanical Engineering and affiliate faculty member of the center, led the Autonomous Ground Vehicles track. Engineering therefore became more than a major listed in a catalog. It became a community organized around shared problems, specialized knowledge, instruments, evidence, and collaboration.
Access to such environments remains uneven. Some high schools maintain extensive robotics laboratories, engineering electives, and research partnerships. Others offer few opportunities beyond the standard curriculum. Talented students may possess the ability to succeed in advanced STEM fields without having encountered the equipment or practices that would reveal that ability. University bridge programs help narrow that gap by providing access to tools, mentors, laboratories, and projects that individual schools may not be able to offer.
American leadership in robotics, artificial intelligence, advanced manufacturing, engineering, and experimental science depends on this long educational pipeline. Graduate laboratories cannot recruit researchers who never had an opportunity to discover their aptitude, and universities cannot expect every engineering student to arrive prepared for open ended technical work. The pipeline must begin before college through serious experiences adapted to the developmental level of motivated students.
SPARC achieved that balance. Students received instruction and guidance, but the machines still had to work. The curriculum was structured, but the engineering problems remained real. Faculty provided support, but students had to assemble, program, test, diagnose, and explain their own systems.
Pascal completed SPARC after an already demanding summer that also included the University of Arizona’s Advanced Astronomy Camp. Arizona immersed him in observational science and astronomy, and Maryland shifted his attention toward engineering, robotics, and autonomous systems. The two experiences shared a deeper foundation. Experimental physics requires observation, measurement, calibration, and interpretation. Robotics requires design, integration, testing, and control. Both demand patience, mathematical reasoning, technical judgment, and respect for evidence.
Programs such as SPARC matter because future scientists and engineers need opportunities to work with real instruments and physical systems before entering college. Students must learn that machines fail, measurements vary, and progress often depends on diagnosing small errors rather than discovering a single dramatic solution. SPARC gave its students structured guidance, access to university laboratories, and responsibility for making their own systems work.
The certificate Pascal received therefore marked more than the completion of a summer program. It represented movement from components to systems, from programming to autonomy, and from interest in engineering to participation in its actual practice. A robot navigating a maze may appear to be a modest achievement, but the habits required to make it succeed reach much farther.
Further Reading
- Maryland Robotics Center
- University of Maryland SPARC Certificate Program
- Autonomous Ground Vehicles Track Syllabus
- 3D Nanoprinting and Additive Manufacturing for Soft Medical Robotics Syllabus
- National Science Foundation: Defining and Empowering the STEM Workforce of Tomorrow
- National Science Foundation: Building the STEM Workforce of the Future