Other services

Robot FPGA and GPU
IP design services

Our design services have developed a pool of expertise and a wealth of robot intellectual property (IP) that solve customer design challenges in robotic areas including sensing, perception, mapping, localization, motion control, low-level control or actuation.

Partner with us!

Why are we the right partner?

Our acceleration kernels and IP solutions are based on proven open-source software, such as the Robot Operating System (ROS) or the Data Distribution Service (DDS), so that you don't spend time redeveloping what already works. Instead, we fork software that has been developed and proven in use by large developer communities. We then add what is missing: higher performance with lower computation times, real-time and power consumption reductions

Our ROS 2 hardware
acceleration methodology

1st step Trace the graph We instrument the core components of the ROS 2 computational graph to research its architecture. 2nd step Benchmark CPU baseline We trace and benchmark the kernels on the CPU to establish a compute baseline which we'll use while creating custom accelerators. 3rd step Accelerate Nodes and the Graph We develop custom compute architectures including soft-cores and acceleration kernels (for both FPGA and GPU) while optimizing the dataflow between Nodes. 4th step Benchmark hardware acceleration We continuously trace and benchmark against the CPU baseline (step 2), speeding up your graph on each iteration until meeting your goals. Start! Accelerated graph
1st step We instrument the core components of the ROS 2 computational grap to research its architecture. 2nd step We trace and benchmark the kernels on the CPU to establish a compute baseline which we'll use while creating custom accelerators. 3rd step We develop custom compute architectures including soft-cores and acceleration kernels (for both FPGA and GPU) while optimizing the dataflow between Nodes. 4th step We continuously trace and benchmark against the CPU baseline (step 2), speeding up your graph on each iteration until meeting your goals. and the Graph Accelerate Nodes Trace the graph Benchmark CPU baseline Benchmark hardware acceleration

Do you have any questions?

Get in touch with our team.

Let's talk