Unmanned Systems Technology 021 | Robot Aviation FX450 l Imaging Sensors focus l UAVs Insight l Liquid-Piston X-Mini l Riptide l Eurosatory 2018 show report l Zipline l Electric Motors focus l ASTS show report

66 Digest | Riptide Autonomous Solutions to take out cost and complexity, and make it a more adaptable architecture at a far lower cost and power level.” The design challenge was to deliver both electrical and hydrodynamic efficiency at the lower cost. He points to sensor makers that are now looking to provide 80% of the capability of their top- end systems at half their original price. But there were also some basic engineering enhancements to the system design. Standardising on one tool and M5 screws for the production version made a major difference to the speed of putting the systems together. That compares to the initial prototype that used M3 screws, as there was a motor with an M3 mount, but then other parts of the system needed 1.5 and 2.5 mm tools. A lot of work also had to be done on the cables and connectors. The wiring harness can add a lot of weight, so the move to Arduino boards with standard connectors helps reduce that considerably. The electronics also used an I2C serial comms link as a bus. While this is low cost and low power, if it breaks then the whole signal chain fails. “We did a lot of work on custom boards for a fault-tolerant I2C bus,” says Smith. “That means we can isolate where the failure is and keep operating.” The main body is built from carbon fibre using a mould for a sailing boat mast, which gives flexibility in the length. 3D printing was used to make the fin. “We wanted better performance than the current 60-100 m; we wanted 200 m,” Smith says. “So after pressure testing the design to failure we adjusted the specification to 300 m. The 3D printing was an ideal way to get complex geometries on parts of the vehicle.” Riptide has now moved to injection moulding those same components to reduce the cost further, but still has the ability to 3D-print components for different features such as a camera or sonar. One issue was about sealing, and how to do it reliably. “When you look at the critical sealing surfaces, these are machined aluminium, really just leaving the 3D-printed components as the hull,” Smith says. “Those sealing rings give us greater strength on the plastic structure.” Riptide publishes the specification of the basic interface rings but doesn’t give the full solid models for the 3D printing. The structure gives other advantages, as transducers and sensors can be bonded into the carbon fibre as special payload rings. A fully aluminium section can be dropped in with transducers for a sonar sensor, for example. That gives a lot of flexibility when building a system for a specific application. It also allows a range of materials to be used, from carbon fibre and glass-filled nylon to aluminium. “When you go to aluminium you get five times the strength, and that change allows us to go down to 1500 m,” Smith says. The company is now looking at 3D-printing titanium rings to allow the UUV to travel down to 6000 m. “There are benefits to smaller vehicles as the walls don’t need to be as thick, and the vehicle cards and battery sections don’t have to change – but you do have to get more length for buoyancy and weight,” Smith says. The other big enabler for deep operation was the efficiency, he says. “Most deep vehicles cost $4-5 million and typically run for 24 hours.” This changes the way the systems are used. “Now it’s a case of having a larger vehicle as a master node for positioning and comms, with a small vehicle that runs deep with a sonar.” August/September 2018 | Unmanned Systems Technology Riptide has evolved its UUV over the past few years, reducing power consumption and improving hydrodynamic efficiency along the way The walls don’t need to be so thick with smaller vehicles – but you do have to get more length for buoyancy and weight