Research
Our current projects examine how control constraints and force feedback shape robotic and haptic interaction. We use benchtop experiments and virtual tasks to test specific system behaviors before considering more complex assistive applications.
Constraint-Based Control for Assistive Robotic Manipulation
Student researcher: Payton Triebold
Faculty mentor: Rohit Singla
Platform: UFactory xArm 7 with Gripper G2
Research Question
How can a robot accommodate requested delivery positions while respecting a defined clearance boundary throughout its motion?
Experimental Approach
This project uses a benchtop spoon-delivery task with a NAO robot serving as a non-human mouth surrogate. An experimental boundary is defined relative to the mouth-center reference and the xArm tool-center point—the reference point used to describe the tool’s position.
The initial study compared two strategies: rejecting commands inside the boundary or redirecting their endpoints to the boundary. Testing covered five requested distances, with five repetitions per distance for each strategy.
Initial Findings
Redirection completed all 25 trials, while blocking permitted completion in 15 of 25 trials. These counts reflect the different command-handling rules: blocking deliberately rejected the ten inside-boundary requests.
The study evaluated completed endpoints, not continuous clearance along the motion path. Endpoint compliance therefore does not establish that the entire movement remained outside the boundary.
Current Development
Current work focuses on recording robot trajectories and evaluating clearance throughout each movement. Planned extensions include force sensing and camera-based localization to support trajectory-aware control. The platform is being developed for benchtop research; safety for human feeding has not been established.
Related presentation: WiSys SPARK Symposium, July 2026
View poster →
Haptic Guidance in Virtual Biopsy Tasks
Student researcher: Troy VerStrate
Faculty mentor: Rohit Singla
Platform: 3D Systems Touch X
Research Question
How do guidance strength and simulated tissue interactions affect depth targeting, force-command behavior, and repeatability in a virtual biopsy task?
Experimental Approach
The Touch X allows an operator to move a virtual tool while receiving force feedback. The project uses this interface to investigate depth targeting and configurable guidance in a simplified biopsy-inspired simulation.
An initial feasibility study compared haptic and non-haptic targeting. Ongoing development introduces simulated tissue layers and adjustable guidance while improving trial control and data recording.
Initial Findings
The feasibility study showed lower depth error during haptic trials and established a working simulation and analysis workflow. Because testing involved one operator and a fixed condition order, the findings are preliminary and do not establish a general benefit across users.
Current Development
Current work focuses on reliable trial recording and characterization of guidance strength, contact stability, and repeatability in a layered simulation. Force logs currently represent commanded rather than independently measured forces, and the simulated layers have not been validated against biological tissue mechanics.
Related presentation: WiSys SPARK Symposium, July 2026
View poster →
Broader Research Interests
Related interests include teleoperation, shared control, redundancy resolution, and dynamical-system modeling. These areas inform our longer-term research direction; the projects above describe the laboratory’s current experimental work.