Unmanned Systems Technology 010 | nuTonomy driverless taxi | Embedded computing | HFE International marine powertrain | Space vehicles | Performance monitoring | Commercial UAV Show Asia report

14 Platform one Nvidia has released details of the Parker chip used on its Drive PX2 embedded processor card for driverless cars (writes Nick Flaherty). The PX2 uses two of the Parker chips, which can run complex neural network software for artificial intelligence. “The design time for a new chip architecture is about three years, so the vision has always been to bring computing capabilities to embedded devices to enter a space that was originally driven by the automotive business,” said Vincent Ngyuen, embedded specialist at Nvidia. “The car industry today does not have enough computing power for what it needs, and the PX2 is a big board – it’s not your average embedded platform.” The Parker consists of two second-generation 64-bit Denver CPU cores (Denver 2.0) paired with four 64-bit ARM Cortex-A57 CPUs. These all work together in a coherent heterogeneous multi-processor configuration to provide up to 1.5 Tflops of performance, compared with 390 Gflops for the previous chip. The Denver 2.0 CPU is a seven-way superscalar processor supporting the ARM V8 instruction set and implements an improved dynamic code optimisation algorithm and additional low-power retention states for better energy efficiency. The two Denver cores and the Cortex-A57 CPU complex are interconnected through Nvidia’s proprietary coherent interconnect fabric. A new 256-core graphics processor in the chip called Pascal is used for the deep learning inference algorithms for self- driving systems. Working with supercomputers in the cloud that also use the Pascal GPU core, Parker-based self-driving cars can be continually updated with newer algorithms and information to improve self-driving accuracy and safety. The chip also supports hardware-enabled virtualisation that supports up to eight virtual machines. This allows car makers to use a single Parker-based Drive PX2 system to concurrently host multiple systems without interference, which is an essential element of the safety process. Volvo and Ford are using the PX2 for their autonomous driving development. Nvidia has also developed a smaller card that uses just one Parker chip, and this is aimed at Advanced Driver Assistant System designs. Driverless cars’ Parker chip Embedded computing October/November 2016 | Unmanned Systems Technology Dr Donough Wilson Dr Donough Wilson is innovation lead at VIVID/ futureVision, which specialises in game- changing thinking for defence, homeland security, and both manned and unmanned aviation innovations. He was first to propose the automatic tracking and satellite download of airliner black box data, technology which is now being adopted. His defence innovations include the automatic cockpit vision system that protects military aircrew from asymmetric high-energy laser attack. As a pilot, he has more than 3000 hours of flying experience in both military and civil environments, and is currently a flying instructor and a flight test examiner. Paul Weighell Paul has been involved with electronics, computer design and programming since 1966. He has worked in the real-time and failsafe data acquisition and automaton industry using mainframes, minis, micros and cloud-based hardware on applications as diverse as defence, Siberian gas pipeline control, UK nuclear power, robotics, the Thames Barrier, Formula One and automated financial trading systems. • Would you be interested in joining our team of consultants? In return for giving us your feedback on some of our articles before they’re published, we’ll highlight your professional expertise in this section of each issue of the magazine. For details contact our Editorial Director Ian Bamsey at [email protected] Unmanned Systems Technology’s consultants Nvidia’s Parker chip for its Drive PX2 card in driverless cars uses two specially designed versions of the ARM core known as Denver