indie has launched the iND881, a next-generation edge artificial intelligence system-on-chip designed to power smart cameras for automotive and robotic applications. The new SoC integrates an AI compute engine with indie's low-latency multi-camera image signal processor to offer an efficient solution for intelligent camera development. This release targets the growing demand for advanced perception systems in safety-critical environments.

Technical Specifications and Architecture

The iND881 is engineered for low power and real-time responsiveness, delivering heterogeneous compute capabilities for edge perception tasks. Its architecture combines a Neural Processing Unit, a versatile Digital Signal Processor, and a quad-core ARM Cortex-A53 CPU. The NPU accelerates AI workloads, while the DSP handles classical signal processing tasks. The CPU supports both industry-standard high-level operating systems and legacy code-base applications.

Applications and Features

The device is particularly beneficial for automotive systems, including advanced driver assistance features such as driver and occupant monitoring and smart eMirrors with blind-spot detection. It also supports smart-camera deployments in robotics and broader physical AI automation applications, including autonomous mobile robots and humanoids, where advanced multi-modal perception is essential.

Key features include a low-latency H.264 encoder for multi-channel video compression and streaming, paired with a high-dynamic-range ISP capable of handling multiple camera inputs with sub-1 ms processing latency. The iND881 supports other sensor modalities, including infrared, thermal, time-of-flight, Radar, and LiDAR. Built on the iND880 platform, the SoC includes a hardware security module for cybersecurity and offers software reuse through a unified software development kit.

Integration and Availability

The iND881 can be combined with emotion3D's production-validated driver and occupant monitoring perception stack to form a single hardware-and-software solution. This integrated approach helps reduce third-party IP complexity and development risk through hardware co-design. The platform can be deployed as a fully pre-integrated package or used independently as a standalone perception software offering.

Fred Jarrar, senior vice president of product line and business management at indie, stated that the launch expands the company's portfolio into edge AI markets such as automotive and humanoid robotics. The iND881 is ASIL-B compliant and automotive qualified. It is sampling now and will be demonstrated at AutoSens and InCabin USA 2026.