Robotics engineer with hands-on experience commissioning mobile robots and manipulators at Fraunhofer IPA. Research focused on learning-based whole-body control using Reinforcement Learning and Behavior Cloning, validated on contact-rich tasks like door opening in MuJoCo. Looking for my next role.
Most mobile manipulators run two independent stacks: a navigation planner drives the base to a pose, then a manipulation planner takes over. The seam between them is where contact-rich tasks fall apart — the arm can't compensate for a base that stopped 5 cm short, the base can't shift to extend the arm's reach.
This research treats the entire mobile manipulator as a single coupled system. BC bootstraps a strong prior from demonstrations; RL refines and recovers when the policy drifts off-distribution. Validated in MuJoCo on door opening — a task that decoupled controllers struggle with by design.
Bring-up and commissioning of AMRs and AGVs — navigation tuning, sensor integration, on-site validation across deployments.
Driver integration, motion planning, and application development for industrial robotic arms across diverse use cases.
Custom ROS2 nodes, launch infrastructure, and integration code purpose-built for each project's robot stack.
Spec → running system: hardware bring-up, software config, debugging, and handover in research-industrial environments.
ROS2 control and path-planning pipeline for mobile robots, with motion-planning algorithms tuned for optimized trajectories. Simulink for system-level design and code validation.
2-axis CNC pen plotter built from scratch on ESP32. Stepper drivers, position sensors, 3D-printed mechanics, firmware in C/C++ — drawing precise designs end-to-end.
Open to robotics, mobile manipulation, robot learning, and applied research roles across the EU. Always happy to talk shop about whole-body control, RL, or a hard engineering problem.
→ dhruvalde.v29@gmail.com