Journal article Open Access

# Novel Arithmetics in Deep Neural Networks Signal Processing for Autonomous Driving: Challenges and Opportunities

Cococcioni, Marco; Rossi, Federico; Ruffaldi, Emanuele; Saponara, Sergio; Benoit Dupont de Dinechin

### Dublin Core Export

<?xml version='1.0' encoding='utf-8'?>
<oai_dc:dc xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
<dc:creator>Cococcioni, Marco</dc:creator>
<dc:creator>Rossi, Federico</dc:creator>
<dc:creator>Ruffaldi, Emanuele</dc:creator>
<dc:creator>Saponara, Sergio</dc:creator>
<dc:creator>Benoit Dupont de Dinechin</dc:creator>
<dc:date>2020-12-24</dc:date>
<dc:description>This article focuses on the trends, opportunities, and challenges of novel arithmetic for deep neural network (DNN) signal processing, with particular reference to assisted- and autonomous driving applications. Due to strict constraints in terms of the latency, dependability, and security of autonomous driving, machine perception (i.e., detection and decision tasks) based on DNNs cannot be implemented by relying on remote cloud access. These tasks must be performed in real time in embedded systems on board the vehicle, particularly for the inference phase (considering the use of DNNs pretrained during an offline step). When developing a DNN computing platform, the choice of the computing arithmetic matters. Moreover, functional safe applications, such as autonomous driving, impose severe constraints on the effect that signal processing accuracy has on the final rate of wrong detection/decisions. Hence, after reviewing the different choices and tradeoffs concerning arithmetic, both in academia and industry, we highlight the issues in implementing DNN accelerators to achieve accurate and lowcomplexity processing of automotive sensor signals (the latter coming from diverse sources, such as cameras, radar, lidar, and ultrasonics). The focus is on both general-purpose operations massively used in DNNs, such as multiplying, accumulating, and comparing, and on specific functions, including, for example, sigmoid or hyperbolic tangents used for neuron activation.</dc:description>
<dc:description>DOI: 10.1109/MSP.2020.2988436</dc:description>
<dc:identifier>https://zenodo.org/record/4564326</dc:identifier>
<dc:identifier>10.1109/MSP.2020.2988436</dc:identifier>
<dc:identifier>oai:zenodo.org:4564326</dc:identifier>
<dc:language>eng</dc:language>
<dc:relation>info:eu-repo/grantAgreement/EC/H2020/826647/</dc:relation>
<dc:relation>doi:10.1109/MSP.2020.2988436</dc:relation>
<dc:rights>info:eu-repo/semantics/openAccess</dc:rights>
<dc:source>IEEE Signal Processing Magazine 38(1) 97-110</dc:source>
<dc:title>Novel Arithmetics in Deep Neural Networks Signal Processing for Autonomous Driving: Challenges and Opportunities</dc:title>
<dc:type>info:eu-repo/semantics/article</dc:type>
<dc:type>publication-article</dc:type>
</oai_dc:dc>

61
204
views