Texas Instruments launches six Arm Cortex vision processors for edge AI applications
Texas Instruments has introduced six vision processors based on Arm Cortex, intending to provide better efficiency for edge AI applications, such as autonomous mobile robots.
Texas Instruments claims that when used with standard software and programming tools, these vision processors enable developers to expand their edge AI designs to various systems.
At the Embedded World 2023, Texas Instruments showcased the AM6xA vision processor, which aims to simplify and decrease the cost of deploying machine vision applications in low-power edge AI systems. The company demonstrated the capabilities of this vision processor in several applications, including retail, industrial robotics and advanced driver assistance system applications.
“In order to achieve real-time responsiveness in the electronics that keep our world moving, decision-making needs to happen locally and with better power efficiency,” says Sameer Wasson, the vice president of processors at Texas Instruments. “This new processor family of affordable, highly integrated SoCs will enable the future of embedded AI by allowing for more cameras and vision processing in edge applications.”
Three new vision processors are available, namely the AM62A, AM68A, and AM69A, powered by the Arm Cortex-A53 or Cortex-A72 central processing unit and a third-generation Texas Instruments image signal processor. These processors feature integrated hardware accelerators capable of delivering edge AI processing for deep learning algorithms, with processing power ranging from 1 to 32 TOPS.
The AM62A vision processors from Texas Instruments are available in three variants–the AM62A3-Q1, AM62A7 and AM62A7-Q1. They support one or two cameras and consume less than 2W of power in vision processing applications. These processors offer a vision processing power of 1 TOPS and are among the most cost-effective options.
The Texas Instruments open-source development tool, Edge AI Studio, will release a public beta version in the second quarter of 2023, allowing users to develop and test AI models. One of its features, the model analyzer, enables remote connection to the hardware to evaluate the model and test its performance on a Texas Instruments embedded processor.
Texas Instruments introduces a 32-bit MCU with integrated analog features
Texas Instruments has developed a new Arm Cortex-M0 microcontrollers portfolio that offers pin-to-pin compatibility with various memory, analog and computing requirements. This approach reduces development costs while maintaining flexibility and performance. The MSPM0L and MSPM0G series are new scalable microcontroller units with customizable options.
The intention of Texas Instruments was to allow developers to dedicate more time to innovation by minimizing the time spent on evaluation and coding. According to the company, this is expected to decrease development time significantly from months to days.
The Arm Cortex-M0 microcontroller portfolio offers a wide range of computing options, ranging from 32 to 80 MHz, with math acceleration and multiple integrated analog signal-chain components configurations.
“TI is building the industry’s most comprehensive portfolio of Arm Cortex-M0+ based MCUs – expanding an already extensive semiconductor offering with options for general-purpose designs,” says Vinay Agarwal, the vice president of MSP Microcontrollers at Texas Instruments. “Our new MCUs provide the flexibility our customers need to enhance the sensing and control capabilities of their systems while cutting cost complexity and design time.”
The Arm Cortex-M0 microcontroller portfolio features an onboard zero-drift operational amplifier that amplifies input signals. This amplifier can automatically adjust the output to eliminate any errors caused by changes in external actors. These amplifiers are ideal for various general-purpose and precision applications requiring signal stability. Such applications include interfaces for precision strain gauges, thermocouples, thermopiles and bridge sensors.
The Texas Instruments LaunchPad development kits for the MSPM0L1306 and the MSPM0G3507 are designed for prototyping embedded applications. These evaluation models come with an onboard debug probe for programming and debugging. Additionally, they feature three buttons, two LEDs, an analog temperature sensor and a light sensor.
Edge computing for the transportation industry
Article Topics
ARM | chip | computer vision | edge AI | IoT | robotics | Texas Instruments | vision processor
Comments