Unmanned Systems Technology 016 | Hydromea Vertex AUV | Power management systems | Unmanned Space Vehicles | Continental CD-155 turbodiesel | Swift 020 UAV | ECUs | DSEI 2017 Show report

October/November 2017 | Unmanned Systems Technology 30 Dossier | Hydromea Vertex AUV Payload integration The current payload – and the single most expensive item aboard the Vertex, at around $20,000 – is the multi-parameter EXO 2 water-quality- measuring sonde from YSI, which Hydromea has modified for integration into the AUV and to which it has added a sensor of its own design. Fitted with up to seven sensor modules at one time, the EXO 2 can measure a wide range of parameters including CTD, ammonium levels, dissolved oxygen content, chloride concentrations, fluorescent dissolved organic matter, nitrates, pH/organic reduction potential, blue-green algae and turbidity. All of them plug into connectors in the front (wet end) of the EXO 2, which itself occupies the payload bay in the nose of the vehicle. The exposed tips of the probes are protected by a 3D-printed plastic frame, while the rear of the wet payload bay features openings to allow water to flow through and past the sensors so that the sensors are always presented with new water to sample. Normally lowered into the water on a cable, the EXO 2 had to be modified for integration into the vehicle, which involved removing the battery pack and tether, and feeding the sonde’s output into the Vertex’s electronics stack. The EXO 2 records all its sensor measurements and makes them available to the vehicle through a serial interface, which allows the information to be used during the mission. With live sensor data, the vehicle can respond to what it senses, and Hydromea recently conducted just such an adaptive sensing experiment in which Vertex created new waypoints to follow up interesting sensor readings. As part of a research project Hydromea is carrying out for the EPFL, the company developed a temperature sensor capable of very high sampling rates and measuring very small temperature changes. The idea is to produce a very high-resolution picture that enables researchers to measure the micro- turbulence caused by bacteria as they swim up towards the light. The sensor is the Mini-FT7 miniature fast temperature profiler, which plugs into the central bay in the EXO 2. Weighing 160 g in air and 100 g in water, it is 25 mm in diameter and 200 mm long. It measures temperatures from -5 to +45 C at a sampling rate of 400 Hz and with a resolution of 20 micro-degrees C. Next steps By the end of this year or early 2018, the two prototype Vertex AUVs will be joined by either five or seven more to allow Hydromea to conduct its first multi-vehicle missions in preparation for offering the capability as a service, with sales of the system to follow later. The first such missions will involve five vehicles, later scaling up to ten or 15 on the way to 20 or more. There is still work to be done on the reliability of individual vehicles, as Bahr points out that however long the mean time between failures is for the multi- vehicle system as a whole it is inevitably divided by the number of vehicles in use. A ‘failure’ might be defined though as anything from the loss of a vehicle to a battery running down unexpectedly quickly, and multi-vehicle systems can lose the odd vehicle without degrading the mission too much. “It’s just a question of having done enough deployments to gain confidence that everything’s fine,” Schill adds, “because you don’t know about the problems you haven’t encountered yet.” For multi-vehicle operations, work still to be done centres on operational procedures and mission control software. The two are intimately linked, and Hydromea is developing its own. “We first used some available drone software, but we very quickly realised it didn’t do everything we needed, such as loading a bathymetric map to get the depth at each point to plan the mission,” Schill says. “Now we are extending that with additional features that make it easier to deal with multiple vehicles simultaneously. “That’s where it really is new – how you define a mission that’s not just defined in terms of waypoints but also in terms of swarm behaviours and formations, and how you define what a formation is supposed to be.” Other key issues include how to programme several vehicles at once instead of one at a time, how to give groups high-level commands that the vehicles can then divide into sub-tasks and parcel out, and how to respond in real time. “We need to keep developing that to see what features we actually need, and then set up a feedback loop from our deployments,” says Schill. Vertex may be a small AUV, but it embodies some big and complex ideas. The company has also developed the Mini-FT7 ultra-fast high-resolution temperature sensor to measure micro- turbulence (Courtesy of Hydromea)