For autonomous grasping by a prosthetic hand, a binarized neural network on FPGA has been investigated, focusing on the trade-off between application quality, FPGA resources, and power consumption. Results have been accepted for publication at the prestigious International Conference on Robotics and Automation, ICRA 2021.

The title of the paper is "Binary-LoRAX: Low-power and Runtime Adaptable XNOR Classifier for Semi-Autonomous Grasping with Prosthetic Hands".

The authors are Nael Fasfous, Manoj-Rohit Vemparala, Alexander Frickenstein, Mohamed Badawy, Felix Hundhausen, Julian Höfer, Naveen-Shankar Nagaraja, Christian Unger, Hans-Jörg Vögel, Jürgen Becker, Tamim Asfour, and Walter Stechele.

The investigation has been conducted in a joint effort between BMW, KIT, and TUM.

ICRA is the IEEE Robotics and Automation Society’s flagship conference and the premier international forum for robotics researchers to present and discuss their work. The 2021 conference will be held from May 30 to June 5, 2021.

http://www.icra2021.org