This post is also available in: עברית (Hebrew)
For vessels operating at sea, avoiding collisions is a basic operational requirement, so collision avoidance is part of operator training. When vessels become highly autonomous, collision avoidance must be incorporated into complex autonomy algorithms and thoroughly tested before the vessels enter the water. For this reason precisely, researchers from the Georgia Tech Research Institute (GTRI), led by Chief Scientist Lora Weiss, have created an assessment tool for systematically stimulating and testing the logic of fully autonomous systems while they are under development before they reach the operational test and evaluation stage.
The technology is called Autonomy Validation, Introspection, and Assessment (AVIA), and was developed with support from the Defense Advanced Research Projects Agency (DARPA) to assess the autonomy logic of unmanned systems.
According to the research institute’s website, AVIA stimulates the actual autonomy logic of the unmanned vessel and can run thousands of assessments faster than in real time and in parallel to study how an autonomous vessel would interact with a dozen or more other vessels. Its graphical user interface allows testers to generate thousands of test scenarios, assess the ability of a vessel to understand a given situation, and determine whether the vessel has responded appropriately.
This allows for an extensive analysis of the full autonomy logic and enables detection of any undesirable behavior earlier in the development phase. Miles Thompson, a GTRI research engineer explained: “It’s very rare to have a collision between two vessels on the open sea today, and we have to make sure that the fact that autonomous vessels have begun emerging equals or improves human performance in this parameter. Using AVIA, we can stimulate the actual autonomy logic of the vessel for thousands of hours to find any issues before the system enters the water for the first time.” Developed for surface and underwater vessels, AVIA could also be useful for evaluating highly autonomous systems designed to operate on the ground, in the air, or even in space.