Three application processor cores from Codasip support RISC-V P extensions

Three 64-bit RISC-V application processor cores have been released by customisable RISC-V processor IP specialist, Codasip. The A70XP provides support for RISC-V P extensions, the A70X‑MP and A70XP‑MP enable the creation of symmetric multi-processor (SMP) systems.

The RISC-V P extension consists of 331 instructions which can be divided into groups. The A70XP includes a single instruction, multiple data (SIMD) unit which executes P extension instructions with single cycle latency. Multi-cycle instructions are pipelined to allow one to be executed every clock cycle. The core can be used for audio encoding/decoding, sensor fusion, computer vision and edge artificial intelligence/machine learning (AI/ML) applications.

The A70X-MP and A70XP-MP cores add multi-core features to the Codasip application processor family, supporting clusters of up to four cores in an SMP configuration. Codasip provides configurable L1 and L2 caches with a scalable microarchitecture. All three application processors use an AXI external interface (Advanced eXtensible Interface, part of the Arm Advanced Microcontroller Bus Architecture 3 and 4 specifications) and support Linux.

The Application RISC-V processors (denoted by product names beginning with A) are based on the same microarchitecture as the A70X (Codasip Bk7). All the Application cores are 64-bit and feature a floating point unit and Atomic instructions. They also support machine, supervisor and user privilege modes and have a memory management unit (MMU), therefore they are able to run Linux. Like other RISC-V cores by Codasip, they have been are developed using Codasip Studio allowing them to be customised to meet domain-specific requirements.

The cores have been developed as a result of combining skills at Codasip’s new design centre in France and Codasip’s main R&D centre in Brno, in the Czech Republic.

The A70X core is available today and the other three cores will be available in the first quarter of 2021.

Karel Masařík, CEO Codasip, said the company expects to introduce more new products in 2021.

Codasip offers two further processor families for the embedded domain, the small and efficient low power embedded processors and more powerful high performance embedded processors. Both of these families are based on the Codasip Bk3 and Bk5 microarchitectures.

Codasip delivers processor IP and design tools, providing IC designers with all the advantages of the RISC-V open ISA, and can customise the processor IP.

The company is a founding member of the RISC-V Foundation and a long-term supplier of LLVM and GNU-based processor solutions, and is committed to open standards for embedded processors.

Codasip was formed in 2014 and is headquartered in Munich, Germany. There are offices in Europe and China, and sales representatives worldwide.

http://www.codasip.com

> Read More

dSpace partners with Microsoft Azure to develop ADAS

Customers of dSpace will have access to the data-driven development of advanced driver assistance systems (ADAS) and autonomous driving (AD) on the Microsoft cloud computing platform Azure, following the announcement that the two companies will work together to offer research and development engineers in the automotive industry an end-to-end solution that is scalable, secure, efficient, and agile.

The development of ADAS/AD functions relies on capturing large volumes of data from the vehicle perception sensors, buses, and networks, and on generating simulated scenarios to analyse the behaviour of the software, individual systems, subsystems as well as complete, integrated systems. The incoming data has to be enriched to a usable format, and it must be easy to distribute to teams working on AI-based development, data replay as well as simulation and validation tasks. This requires a powerful, flexible, and centralised data storage system, as well as a scalable and computational infrastructure with AI- and machine learning (ML) -based tools that can run seamlessly in the same environment.

Supplier to the automotive industry, dSpace offers mature and proven end-to-end solutions for ADAS/AD simulation and validation, including data logging, data enrichment, advanced simulation models, and data management software. Microsoft’s global, open, and scalable cloud platform allows businesses meet security, privacy and compliance requirements while innovation and development continues.

In this way, says dSpace, an integrated end-to-end solution for data-driven development can be achieved, allowing automotive OEMs and suppliers to focus on algorithm development.

“We combine our comprehensive and mature simulation portfolio with the highly scalable computational infrastructure of Microsoft Azure so that our customers can take full advantage of best-in-class solutions of both areas,” says Tino Schulze, executive vice president of Automated Driving & Software Solutions at dSpace.

“Collaborating with Microsoft will enable us to further expand our expertise in the areas of cloud computing and big data. Together, we will tackle the challenges of our automotive customers, supporting them in getting self-driving cars on the road faster,” explained Martin Goetzeler, CEO of dSpace.

dSpace provides solutions for developing connected, autonomous, and electrically powered vehicles. Automotive manufacturers and their suppliers use the company’s end-to-end solution range to test the software and hardware components of their new vehicles long before a new model is allowed on the road. Engineers also rely on dSpace expertise in aerospace and industrial automation. Its portfolio ranges from end-to-end solutions for simulation and validation to engineering and consulting services as well as training and support.

The company is headquartered in Paderborn, Germany and has three project centres in Germany. Customers are served through regional companies in the USA, the UK, France, Japan, China, and Croatia.

http://www.dspace.de

> Read More

Aaeon joins with Lips to accelerate Nvidia’s Isaac AI

Industrial vision and robotic applications are the target applications for an Nvidia Isaac certified combination of Aaeon’s Boxer-8240AI (powered by Nvidia Jetson AGX Xavier), with Lips’ Lipsedge AE400 industrial 3D camera.

Nvidia Isaac software development kit (SDK) is believed to be the industry’s first robotic AI development platform to include simulation, navigation and manipulation. These features help Isaac SDK users to quickly program, train, fine tune, revise and deploy industrial AI. With Nvidia Isaac certification, the Boxer-8240AI with LIPSedge AE400 camera can help power and accelerate the development of smart factory applications, for example autonomous guided vehicles (AGV) and vision guided robots.

The Boxer-8240AI embedded AI edge box PC’s Nvidia Jetson AGX Xavier features the 512-core Volta graphics processor unit (GPU) with 64 Tensor cores, capable of AI processing speeds up to 32Terra operations per second. The Boxer-8240AI is rugged and fanless design to operate in almost any industrial environment, with an I/O load including four Gigabit LAN ports, which are particularly useful for industrial robotics and autonomous vehicles and robots, notes Aaeon.

The LIPSedge AE400 Industrial 3D camera features a rugged enclosure and IP67 classification, making it equally suitable for stringent industrial environments. It supports Gigabit Ethernet with PoE, allowing for faster data transfer and flexibility in deployment. The camera has been used extensively in industrial applications including 3D vision guided robots and AGV/AMR deployment. Lips also provides developers a suite of software tools to help build inferences and applications and reduce deployment time.

Lips provides AI-enabled 3D sensing. It designs, builds, and customises 3D depth cameras, middleware, and provide turnkey solutions. It delivers customised machine vision and AI for multiple industries.

Aaeon was established in 1992 and designs and manufactures industrial IoT and AI edge solutions.

Aaeon provides industrial motherboards and systems, rugged tablets, embedded AI Edge systems, uCPE network appliances, and LoRaWAN/WWAN products. It also works closely with cities and governments to develop and deploy smart city ecosystems and with premier chip designers to deliver stable, reliable platforms, and is recognized as an Nvidia Preferred Partner.

http://www.aaeon.com

> Read More

Palm-sized industrial motherboard adds intelligence to machine vision

The AIMB-U233 is a palm-sized motherboard by Advantech that is targeted at smart applications. It is the first Core i platform in Advantech’s UTX family.

The small form factor design is powered by powerful 8th generation Intel Core processors (i3-8145UE / i5-8365UE / i7-8665UE). The AIMB-U233 motherboard combines high computing performance with diverse I/O ports and M.2 expansion slots for digital applications that require a small footprint, such as integrating into kiosks as well as medical and factory automation applications.

Smart factory applications require enhanced computing power integrated peripheral devices to enable machine vision and deep learning yet have limited installation spaces. The Advantech AIMB-U233 motherboard is compact (112 x 137mm) and offers four COM, four USB, two HDMI, two GbE LAN, 16-bit general purpose input/output (GPIO) and three M.2 expansion slots. The Intel Core i7-8665UE processor is particularly suitable for complex computing tasks, says Advantech.

The M.2 expansion slots allow increased functionality and flexible configuration options. The AIMB-U233 is equipped with M.2 M-key, M.2 E-key, and M.2 B-key compatibility to support diverse applications and boost computing performance. Combining the M.2 M-key interface with four-lane NVMe PCIe can increase storage read/write speeds up to 400 per cent compared to a traditional SATA 3.0 SSD. The system can automatically switch into SATA mode when a cost-effective SATA SSD is installed.

To meet the increasing demands of wireless connectivity, the M.2 E-key and M.2 B-key connectors can be used to integrate Wi-Fi/Bluetooth and 4G LTE modules and support a wide range of wireless applications. Both expansion slots can also be used for storage via a two-lane PCIE SSD (M.2 E-key) and/or SATA SSD (M.2 B-key). For machine vision applications in automated factories, the M.2 E-key connector supports the use of Advantech Movidius modules for artificial intelligence (AI) computing.

For security and to protect against cybersecurity attacks, the AIMB-U233 motherboard is equipped with TPM 2.0 to enable data security with hardware-based encryption and authentication. The provision of RAID support enables data mirroring to mitigate the impact of hardware failures.

Software options include the over the air (OTA) BIOS, offering seamless updates and a backup recovery mechanism to reduce the need for servicing by equipment technicians.

The board supports a number of Linux distributions and offers a Linux image based on Ubuntu 20.04, a part of the Windows 10 IoT.

To resolve minor computing glitches, the embedded USB controls can be used to discontinue power supply from the device. The COM ports can be configured to RS-232/422/485 mode via the BIOS, which is KVM-accessible through Intel vProtm.

http://www.advantech.eu

> Read More

About Smart Cities

This news story is brought to you by smartcitieselectronics.com, the specialist site dedicated to delivering information about what’s new in the Smart City Electronics industry, with daily news updates, new products and industry news. To stay up-to-date, register to receive our weekly newsletters and keep yourself informed on the latest technology news and new products from around the globe. Simply click this link to register here: Smart Cities Registration