Blog
Tensor and Arm Compute Collaboration to Create Personal Robocar
Tensor and Arm announced a multi-year strategic collaboration to develop a foundational compute architecture to create the first agentic AI personal Robocar. Tensor will implement Arm’s compute platform.
The effort will unify hardware and software to drive physical AI workloads throughout the entire vehicle. More than 400 safety-capable, power-efficient Arm-based cores are in each vehicle. The companies will work together to enable the Level 4 autonomous capabilities required for next-generation mobility.
Leveraging Arm’s software ecosystem, the Robocar is powered by a vertically integrated Level 4 autonomy stack and comprehensive sensor suite. The suite includes 37 cameras, 5 LiDARs, 11 radars, 22 microphones, 10 ultrasonic sensors, 3 IMUs, GNSS, 16 collision detectors, 8 water-level detectors, 4 tire-pressure sensors, and a smoke detector.
The hardware and software integration, supported by triple-channel 5G connectivity, represents a meaningful advancement in automotive sensing, enabling continuous perception, environmental awareness, and resilient system performance in a variety of operating conditions. Tensor is using the Arm compute platform to distribute safety-capable intelligence across the vehicle so that the Robocar can safely perceive and navigate its environment.
Among the Tensor Robocar’s 433 Arm-based cores are the Neoverse AE for high-throughput AI processing; Cortex-X for agentic AI cabin and system control; Cortex-A for drive-by-wire, LiDARs, redundancy, and general compute; Cortex-R for real-time safety-critical systems; and Cortex-M for low-power subsystem management.
The Arm compute platform enables Tensor to deploy AI workloads across diverse compute domains while meeting automotive safety, thermal, and power requirements. It operates in concert with NVIDIA-accelerated AI processing to support Tensor’s proprietary autonomy stack.
