This paper discusses the introduction of an integrated Posit Processing Unit (PPU) as an alternative to Floating-point Processing Unit (FPU) for Deep Neural Networks (DNNs) in automotive applications. Autonomous Driving tasks are increasingly depending on DNNs. For example, the detection of obstacles by means of object classification needs to be performed in real-time without involving remote computing. To speed up the inference phase of DNNs the CPUs on-board the vehicle should be equipped with co-processors, such as GPUs, which embed specific optimization for DNN tasks. In this work, we review an alternative arithmetic that could be used within the co-processor. We argue that a new representation for floating point numbers called Posit is particularly advantageous, allowing for a better trade-off between computation accuracy and implementation complexity. We conclude that implementing a PPU within the co-processor is a promising way to speed up the DNN inference phase.
Exploiting Posit Arithmetic for Deep Neural Networks in Autonomous Driving Applications
2018-01-01
Conference paper
Electronic Resource
English
intelligent transportation system , mobile robot , neural net , road vehicle , Posit arithmetic , Deep Neural Network , Floating-point Processing Unit , automotive application , Autonomous Driving task , DNN inference phase , autonomous Driving , Posit Processing Unit , floating point number , Hardware , Sensor , Open area test site , Autonomous vehicle , Task analysi , Planning , Software
DDC: | 629 |
PATCHING DEPLOYED DEEP NEURAL NETWORKS FOR AUTONOMOUS MACHINE APPLICATIONS
European Patent Office | 2022
|An Autonomous Driving Approach Based on Trajectory Learning Using Deep Neural Networks
Springer Verlag | 2021
|British Library Conference Proceedings | 2017
|