English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Perception and control systems for autonomous vehicles are an active area ofscientific and industrial research. These solutions should be characterised byhigh efficiency in recognising obstacles and other environmental elements indifferent road conditions, real-time capability, and energy efficiency.Achieving such functionality requires an appropriate algorithm and a suitablecomputing platform. In this paper, we have used the MultiTaskV3detection-segmentation network as the basis for a perception system that canperform both functionalities within a single architecture. It was appropriatelytrained, quantised, and implemented on the AMD Xilinx Kria KV260 Vision AIembedded platform. By using this device, it was possible to parallelise andaccelerate the computations. Furthermore, the whole system consumes relativelylittle power compared to a CPU-based implementation (an average of 5 watts,compared to the minimum of 55 watts for weaker CPUs, and the small size (119mmx 140mm x 36mm) of the platform allows it to be used in devices where theamount of space available is limited. It also achieves an accuracy higher than97% of the mAP (mean average precision) for object detection and above 90% ofthe mIoU (mean intersection over union) for image segmentation. The articlealso details the design of the Mecanum wheel vehicle, which was used to testthe proposed solution in a mock-up city.

Implementation of a perception system for autonomous vehicles using a detection-segmentation network in SoC FPGA