This example demonstrates the capabilities of the S&R-rescue methods investighated in Two!Ears. To that end, a baseline visualizer (LVTE) shows the current scenario status in 2 dimensions from a bird's view. The robot is indicated as a blue star with a short line sketching the heading direction of the robotic device. Sound sources are displayed as red crosses, when they become active, they are enveloped with a blue circle. The current task of the robot is displayed in the figure's header, along with the current system time. Note: no actual audio processing is done in this example, this is left to the more powerful BEFT. The robot behaves as reported in Deliverable 4.3, chapter 5, section 4 (please refer to this chapter for more information on the given scenario). While moving, the robot hypothesizes the positions of each accessible sound source, marking them with green asterisks. Once all sources have been locked, and the situation turns into a catastrophe setting, the robot starts to rescue all animate entities from the scene, as described in D4.3, section 5.4. The knowledge sources used in this S&R setting are the BindingKS, AuditoryObjectFormationKS, AuditoryMetaTaggingKS, HazardAssessmentKS, PlanningKS, which are described in detail in Deliverable 4.3, sections 5.2 and 5.3. A more technical overview of these KSs can also be found in Deliverable 6.1.3, section 3.4.4. Please note that the UpdateEnvironmentKS also used here was already described in Deliverable 4.2, section. 2.2.1. Especially the PlanningKS can be modified to alter the robots behavior in the given scenario.
NOTE: the Deliverables can be found at http://twoears.eu/publications/
Just start the example by typing 'searchAndRescue'. From here on, everything is automated: the scene is loaded from the given environment descriptor, the appropriate sound source activation schedule is loaded, and the simulation is started. You will then see the virtual robot from a bird's view perspective, performing the search-and-rescue task described in Deliverable 4.3, section 5.4. Note that the given scene is actually a proof of concept, following strictly the discussions in Deliverable 4.3. To set up new scenarios and/or employ baseline sound processing with the SSR auralizer, please contact [email protected] for potential use of the more powerful BEFT, which, however, requires a non-trivial installation procedure.