Imagination Technologies announced PowerVR Automotive, a product package designed to enable growth and acceleration of the automotive semiconductor industry. Combining five elements—hardware IP, software, tools, documentation, and long-term support—PowerVR Automotive enables existing customers and new entrants in the market to meet the safety standards and performance requirements of the autonomous driving industry

Imagination is introducing the PowerVR Series8XT-A GPU core to complement its existing IP already used extensively in infotainment, digital cockpit, and advanced driver-assistance systems (ADAS). PowerVR Series8XT-A has enhanced recovery and reliability features to make it easier for system-on-chip (SoC) designers to obtain automotive safety certification for their SoCs aimed at digital human-machine interfaces (HMIs) and autonomous vehicles (AVs).

PowerVR Automotive provides customers with support documentation, automotive safety, and traceability information that is designed to ensure customers’ SoCs can achieve the ASIL-levels of functional safety, meeting industry standards such as ISO 26262 and IEC 61506 at the system level.

Based on the PowerVR Furian architecture, the Series8XT-A is a new member of the Series8XT family of GPUs targeted at cluster, HUD, infotainment, and the digital cockpit, as well as ADAS functions. The Series8XT-A provides power, performance, and area (PPA) metrics and enables new features in hardware and software for fault detection and recovery of the GPU. Other new features include Error-Correcting Code (ECC), Logic Built-in Self-test (LBIST), as well as industry-leading verification strategies.

Featuring unique built-in hardware virtualization, Series8XT-A enables automotive OEMs to keep multiple applications and services running securely in isolated containers, with guaranteed quality of service (QoS) for each.

The PowerVR Series2NX Neural Network Accelerators (NNA) are showcased through ADAS applications such as driver monitoring, driver gaze tracking, seat occupancy, road-sign detection, drivable path analysis, road user detection, and driver recognition.