Papers
2
Total Citations
17
H-Index
2
About
Win-San Khwa is a leading researcher in energy-efficient edge computing and hardware acceleration for autonomous micro-robotics. His work centers on developing specialized accelerator architectures that integrate embedded resistive RAM (RRAM) with dedicated compute engines, enabling real-time perception and localization on millimeter-scale platforms. Khwa’s most notable contribution is the design of a 40nm VLIW edge accelerator featuring 5 MB of ultra-low-power RRAM (0.256 pJ/bit) and a dedicated localization solver, specifically optimized for bristle robot surveillance. This accelerator efficiently handles both the neural network inference stack for perception and state-space equation solving for localization, achieving unprecedented energy efficiency in a compact form factor. His 2024 paper on this system has garnered 15 citations, highlighting its immediate impact on the field. By addressing the critical challenge of balancing compute capability with extreme energy and size constraints, Khwa’s work enables practical autonomous surveillance in tiny robots, pushing the boundaries of what is possible in edge AI and microrobotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2