Lewis Boyd
Papers
2
Total Citations
6
H-Index
2
About
Lewis Boyd is a robotics researcher whose work lies at the intersection of deformable object manipulation and sensor-efficient control. His primary research areas include physics property prediction for soft materials and deep reinforcement learning for hand-eye coordination. Boyd’s most notable contribution is the development of a Physics Similarity Neural Network, a novel approach that enables robots to predict the physical parameters of fabrics and garments without direct measurement—a notoriously difficult challenge in robotic manipulation. This work, published in 2022, has already garnered 4 citations for its practical implications in automating garment handling and soft object interaction. In earlier work, Boyd explored deep reinforcement learning for hand-eye coordination, introducing a "software retina" concept that reduces the computational burden of processing raw pixel inputs. His 2020 study demonstrated that agents could achieve coordinated control with significantly fewer training steps, addressing a critical bottleneck in vision-based robotic learning. Boyd’s research is particularly relevant for industries seeking to automate tasks involving flexible materials, from textile manufacturing to assistive robotics, and his innovative approaches to reducing hardware demands make his work both practical and forward-thinking.
Research Focus
Key Achievements
Top Papers
- 1
- 2