Unmanned Systems Technology 018 | CES show report | ASV Global C-Cat 3 USV | Test centres | UUVs insight | Limbach L 275 EF | Lidar systems | Heliceo DroneBox | Composites

16 February/March 2018 | Unmanned Systems Technology Platform one | CES Production of Platform 3.0 vehicles begins this spring from stock Lexus LS models. Some of them will be built with dual cockpit control for testing TRI’s Guardian software for automated driving, experimenting with effective methods of transferring vehicle control between a human test driver and the automated system while maintaining a safety driver as a back-up. Aquantia used the show to launch a series of multi-gigabit Ethernet chips for autonomous vehicles. It is working with Nvidia to provide these high-speed links on the Pegasus and Xavier platforms. The AQcelerate Automotive devices provide 10 Gbit/s data rates as a physical (PHY) controller, a media access controller (MAC) and a combined MAC+PHY Ethernet controller using the PCIe format “Vehicle sensors and cameras collect huge amounts of data that need instant processing to allow the vehicle to make critical decisions to ensure the safety of the passengers and anyone sharing the road,” said Faraj Aalaei. “Moving data between the sensors and the compute nodes within the in-vehicle network requires multi-gigabit capabilities to provide safe and secure driving.” The chips are aimed at Level 4 and Level 5 self-driving vehicles that need new network architectures for high- bandwidth links for the growing number of sensors and high-resolution cameras, alongside redundancy of all the function- critical components and systems to provide high levels of safety. Aquantia is also working with connector maker Molex on high-reliability 10 Gbit/s automotive Ethernet network connectors. Laser sensor technology company LeddarTech has teamed up with virtual prototyping company Optis to enable the industrial simulation of advanced Lidar systems. “With our LeddarCore System-on-Chip, Tier 1 manufacturers and integrators can build highly optimised Lidar sensors to OEMs’ specific requirements. Optis’ simulation software fits perfectly within that development model,” said Michael Poulin at LeddarTech. The deal allows OEMs and suppliers to integrate and test the next generation of Lidar technology developed around the LeddarCore chip before release of the hardware. All the capability is simulated in a computer environment to allow complex tests to be performed. The test code can then be run on the hardware when it is ready to confirm that the system performs correctly. The simulation software from Optis virtually recreates cameras and Lidar operations on autonomous cars and simulate operations in real-life scenarios. Using light and laser modeling tools, Optis helps in the detailed design of LeddarCore-based Lidar systems. With real material measurements provided by LeddarTech, the Optis software can validate the Lidar model and simulate the correct response from the Lidar in real-time through a virtual closed- loop simulation with automated driving functions. “With this partnership with Optis, automotive companies can now virtually implement, test and validate our Lidar technology very early in the design process. This also lets them fine-tune their system at a more global scale and define the specific configurations, should they require customised Lidar solutions,” Poulin added. Chinese company Baidu announced its second-generation technology for driverless cars at the show, combining systems and software from multiple suppliers in a single, open stack. Apollo 2.0 makes all four of the platform’s major modules – cloud services, software platforms, reference hardware platforms and reference vehicle platforms – open source. It has more than 165,000 lines of code, and Aquantia’s technology is aimed at enabling multi-gigabit networking for Level 4 and 5 autonomy Baidu’s Apollo 2.0 driverless car technology is now open source