.png){kind=logo}
Haoran Yang
Entry-level Robotics Research Assistant specializing in contact-implicit MPC and manipulation
Connect with Haoran
Haoran already has a relationship with Reval, so a warm intro from us gets a much better response than cold outreach.
Recommended
Already have an account?
About
Robotics software engineer who built and tuned a contact-implicit MPC controller for a full planar pushing manipulation pipeline (“Push Anything”), including a key fix for complementarity violations that eliminated “ghost pushes” and cut time-to-goal from 40s to 25s. Hands-on with ROS/MoveIt on real robot pick-and-place, improving hardware grasp reliability through TF/frame debugging, and uses Drake/URDF for simulation, contact detection, and MPC development.
Experience
Education
Key Strengths
- Designed and implemented contact-implicit MPC for planar pushing manipulation
- Debugged and fixed complementarity-violation failure mode (“ghost pushes”) by aligning QP and projection steps and adding penalties
- Demonstrated measurable performance improvement (C-shaped object time-to-goal reduced 40s to 25s)
- Balanced planning horizon vs control rate in real-time MPC tuning (selected N=12 after experimentation)
- ROS/MoveIt integration with real hardware; improved grasp reliability via TF debugging and frame alignment
- Adapted existing 3D manipulation codebase to planar pushing by rewriting functions and communication ports
Discover more candidates like Haoran
Search across thousands of pre-screened, high-quality, high-intent candidates on Reval.
Search Talent