Testing the Waters: Design of Replicable Experiments for Performance Assessment of Marine Robotic Platforms

摘要

Robotics is becoming part of our daily life through home automation (called domotic) and assistive applications that are taking place in habitations, up to industrial and service employments. With this strict cooperation between humans and robots, it is of absolute necessity to assess the robotic systems' capabilities and performance to ensure safety and reliability according to actual and significant criteria and parameters. Within this context, there are delays in technology transfer from research frameworks toward actual applicative scenarios. These delays are caused by the lack of methodologies and standardized procedures for the experiment execution and result comparison. A further restraint to the needed process of technology transfer is due to the inability to understand the actual capabilities of the systems and to be aware and confident (or not) about what they can be realistically employed for. In many cases, this can also turn into the inability to define specific regulations for the employment of robotic platforms to comply with the current laws. This is the case for marine robotics. Tethered vehicles [e.g., remotely operated vehicles (ROVs)] are considered part of the vessel equipment when they are operated from a ship (because of the physical connection provided by the cable). Conversely, whenever autonomous systems, such as unmanned surface vehicles (USVs) or autonomous underwater vehicles are commanded through an acoustic or Wi-Fi/radio communication link, they are considered as navigating platforms and are subject to the regulation in force. For the commercial employment of these vehicles, it is then necessary to strongly assess the vehicle's capabilities in a standardized way, to guarantee the operating performance and evaluate limitations of the target system.