Iterator-Based Temporal Logic Task Planning

Abstract

Temporal logic task planning for robotic systems suffers from state explosion when specifications involve large numbers of discrete locations. We provide a novel approach, particularly suited for tasks specifications with universally quantified locations, that has constant time with respect to the number of locations, enabling synthesis of plans for an arbitrary number of them. We propose a hybrid control framework that uses an iterator to manage the discretised workspace hiding it from a plan enacted by a discrete event controller. A downside of our approach is that it incurs in increased overhead when executing a synthesised plan. We demonstrate that the overhead is reasonable for missions of a fixed-wing Unmanned Aerial Vehicle in simulated and real scenarios for up to 700 000 locations.

Assured Mission Adaptation of UAVs

“…One potential limitation of the approach is that discrete event controller synthesis may not scale to large missions. However, techniques such as Iterator-Based Planning [34] have shown that missions for hundreds of thousands of discrete locations are tractable. Although we extended DCU for the purpose of typical recurring requirements in robotic missions, a comprehensive study of extensions to DCU to account for more complex liveness requirements is still needed and would be interesting to develop in future work.…”

“…The SITL simulator allows us to test UAV systems loaded with ArduPilot firmware (e.g., custom made as in [33] or commercial as the 3DR Solo Drone [31]) without the UAV hardware. This simulator (see [34]) has been used to seamlessly go from testing to actually flying a custom made fixed-wing vehicle (e.g., [35]) based on a Pixhawk (e.g., [36], [37]).…”

“…In large discrete regions, cover missions with explicitlocation modelling do not scale well as one fluent for every location is needed to track if it has been covered. Instead, we use an alternative discrete abstraction strategy: Iterator-Based Planning [34]. This follows the idea of sensor-based planning (e.g., [27]) in which a sensor (i.e., code) is provided to identify if a location correspond to the area to be covered.…”

“…The controller immediately does stopOld but cannot do startNew as Reset does not hold (see Θ 3 ). Further, to do reset it must not be moving (see Iterator requirements [34]), thus it first waits until the new location is reached (which it can assume will eventually happen), then resets the iterator and then does startNew and reconfig. Between startNew and reconfig, the update controller chooses to do a spin (do.spin) since it does not violate any requirement.…”

Ontology-based task planning for autonomous unmanned system: framework and principle

“…Ontology-based approaches for autonomous systems Knowledge is the ladder to human wisdom and also an important basis for unmanned systems to improve their autonomous ability. The task planning of unmanned systems is not only a heavy workload but also requires the support of domain knowledge [15]. At present, the ontology approach has received extensive attention in the field of robotics, which has realized multiple robot ontologies and has been applied in the field of task planning.…”