Papers
18
Total Citations
456
H-Index
11
About
Brian Plancher is a robotics and computing researcher whose work spans robot motion planning, hardware acceleration, and accessible robotics education. He is best known for bridging the gap between computational efficiency and real-world robotic control, with a particular focus on making sophisticated algorithms viable on resource-constrained platforms. His highly cited 2017 paper on MIT's Beaver Works Summer Institute (104 citations) introduced an influential high school robotics curriculum centered on self-driving race cars, demonstrating his commitment to broadening STEM engagement. On the algorithmic side, his work on Constrained Unscented Dynamic Programming (54 citations) advanced trajectory optimization for underactuated robots, while subsequent research on parallel differential dynamic programming explored GPU-based acceleration for these computationally intensive methods. Plancher's "Robomorphic Computing" framework (42 citations) pioneered robot-morphology-parameterized hardware accelerators, a concept extended through RoboShape and RobotCore to enable scalable, open-architecture acceleration within ROS 2 ecosystems. His TinyMPC project (40 citations) brought model-predictive control to microcontrollers, and his work on tiny robot learning (38 citations) charts a roadmap for deploying machine learning on low-cost autonomous platforms. Collectively, his research has reshaped how the robotics community thinks about co-designing algorithms, hardware, and education.
Research Focus
Key Achievements
Top Papers
- 1
- 2Constrained unscented dynamic programming54 citations · 2017
- 3
- 4TinyMPC: Model-Predictive Control on Resource-Constrained Microcontrollers40 citations · 2024
- 5Accelerating Robot Dynamics Gradients on a CPU, GPU, and FPGA39 citations · 2021
- 6
- 7A Performance Analysis of Parallel Differential Dynamic Programming on a GPU31 citations · 2020
- 8
- 9RobotCore: An Open Architecture for Hardware Acceleration in ROS 222 citations · 2022
- 10