Available for the CEVA-XM family of intelligent vision DSPs and the NeuPro family of AI processors, CEVA-SLAM incorporates the hardware, software and interfaces required to lower the entry barrier for companies looking to integrate efficient SLAM implementations into low-power embedded systems.
The CEVA-SLAM SDK is designed to accelerate SLAM-based application development by incorporating the hardware, software and interfaces required to enable an efficient SLAM implementation into any embedded system.
The SDK includes a detailed interface from a CPU to offload the heavy lifting SLAM blocks to the CEVA-XM DSP. These building blocks utilise the DSP efficiency to support both fixed point and floating point math and extend the device’s battery life.
The SDK building blocks include capabilities for image processing (including Feature Detection, Feature Descriptors, Feature Matching), Linear Algebra (including Matrix Manipulation, Linear Equation Solving), Fast Sparse Equation solving for Bundle Adjustment and more.
Running a full SLAM tracking module on the CEVA-XM6 DSP at 60 frames per second consumes only 86mW, CEVA says. When deployed with a CEVA-XM DSPs or a NeuPro AI processor, customers can address a range of use cases and applications requiring SLAM, such as visual positioning, along with classical and Neural Network workloads for imaging and vision, all in a unified hardware platform designed to be easily programmed.
SLAM is the underlying technology that enables high-accuracy 3D mapping of a device’s surroundings, according to CEVA, and is a key component for a range of emerging devices, including AR/VR headsets, drones, robots and other autonomous machines.
