Archytas: A Framework for Synthesizing and Dynamically Optimizing Accelerators for Robotic Localization
Weizhuang Liu, Yu Bo, Yiming Gan, Qiang Liu, Jie Tang, Shaoshan Liu, Yuhao Zhu
Abstract
Despite many recent efforts, accelerating robotic computing is still fundamentally challenging for two reasons. First, robotics software stack is extremely complicated. Manually designing an accelerator while meeting the latency, power, and resource specifications is unscalable. Second, the environment in which an autonomous machine operates constantly changes; a static accelerator design leads to wasteful computation.
Topics & Concepts
Computer scienceComputationRoboticsSoftwareRobotEmbedded systemLatency (audio)Artificial intelligenceDistributed computingSoftware engineeringOperating systemProgramming languageTelecommunicationsModular Robots and Swarm IntelligenceRobotics and Sensor-Based LocalizationRobotic Path Planning Algorithms