Thesis Defense: A Design Methodology for Computer Architecture Parameterized by Robot Morphology
Speaker
Sabrina M. Neuman
CSAIL-MIT
Host
Professor Srinivas Devadas
CSG - CSAIL - MIT
Abstract:
Robots that safely interact with people are a promising solution to address societal challenges from elder care to hazardous work environments. A key computational barrier to the robust autonomous operation of complex robots is running motion planning online at real-time rates. A performance gap of at least an order of magnitude has emerged: robot joint actuators respond at kHz rates, but promising online optimal motion planning for complex robots is limited to 100s of Hz by state-of-the-art software. While domain-specific hardware accelerators have improved the performance of other stages in the robotics pipeline such as perception and localization, relatively little work has been done for motion planning. Moreover, designing a single accelerator is not enough. It is essential to map out design methodologies to keep the development process agile as applications and robot platforms evolve.
I address these challenges by developing a generalized design methodology for domain-specific computer architecture parameterized by robot morphology. I (i) describe the design of a domain-specific accelerator to speed up a key bottleneck in optimal motion planning, the rigid body dynamics gradient, which currently consumes up to 90% of the total runtime for complex robots. Acceleration is achieved by exploiting features of the robot morphology to expose fine-grained parallelism and matrix sparsity patterns. I (ii) implement this accelerator on an FPGA for a manipulator robot, to evaluate the performance compared to existing CPU and GPU solutions. I then (iii) generalize this design to prescribe an algorithmic methodology to design such accelerators for a broad class of robot models, fully parameterizing the design according to robot morphology. This research introduces a new pathway for cyber-physical design in computer architecture, methodically translating robot morphology into accelerator morphology. The motion planning accelerator produced by this methodology delivers a meaningful speedup over off-the-shelf CPU and GPU hardware. Shrinking the motion planning performance gap will enable roboticists to explore longer planning horizons and implement new robot capabilities.
Thesis Supervisor: Srini Devadas
Thesis Committee: Srini Devadas (MIT), Daniel Sanchez (MIT), Vijay Janapa Reddi (Harvard University)
----------------------------------
*Zoom Meeting: Please contact Srini Devadas at devadas at mit dot edu for a meeting link.
Robots that safely interact with people are a promising solution to address societal challenges from elder care to hazardous work environments. A key computational barrier to the robust autonomous operation of complex robots is running motion planning online at real-time rates. A performance gap of at least an order of magnitude has emerged: robot joint actuators respond at kHz rates, but promising online optimal motion planning for complex robots is limited to 100s of Hz by state-of-the-art software. While domain-specific hardware accelerators have improved the performance of other stages in the robotics pipeline such as perception and localization, relatively little work has been done for motion planning. Moreover, designing a single accelerator is not enough. It is essential to map out design methodologies to keep the development process agile as applications and robot platforms evolve.
I address these challenges by developing a generalized design methodology for domain-specific computer architecture parameterized by robot morphology. I (i) describe the design of a domain-specific accelerator to speed up a key bottleneck in optimal motion planning, the rigid body dynamics gradient, which currently consumes up to 90% of the total runtime for complex robots. Acceleration is achieved by exploiting features of the robot morphology to expose fine-grained parallelism and matrix sparsity patterns. I (ii) implement this accelerator on an FPGA for a manipulator robot, to evaluate the performance compared to existing CPU and GPU solutions. I then (iii) generalize this design to prescribe an algorithmic methodology to design such accelerators for a broad class of robot models, fully parameterizing the design according to robot morphology. This research introduces a new pathway for cyber-physical design in computer architecture, methodically translating robot morphology into accelerator morphology. The motion planning accelerator produced by this methodology delivers a meaningful speedup over off-the-shelf CPU and GPU hardware. Shrinking the motion planning performance gap will enable roboticists to explore longer planning horizons and implement new robot capabilities.
Thesis Supervisor: Srini Devadas
Thesis Committee: Srini Devadas (MIT), Daniel Sanchez (MIT), Vijay Janapa Reddi (Harvard University)
----------------------------------
*Zoom Meeting: Please contact Srini Devadas at devadas at mit dot edu for a meeting link.