Motion planning with temporal logic tasks and constraints

Add to your list(s) Download to your calendar using vCal

• Jana Tumova, Assistant Professor at KTH Stockholm
• Friday 04 June 2021, 10:00-11:00
• https://bham-ac-uk.zoom.us/j/83482525060?pwd=V2NqMDVKaXVTWWZoR2VTUXA5MmRmUT09.

If you have a question about this talk, please contact Martin Rudorfer.

Temporal logics are a powerful tool allowing to express involved planning objectives, such as periodic surveillance, sequencing, or request-response tasks, as well as various motion spatial and temporal constraints and preferences. In this talk, we discuss how we can enhance motion planning algorithms to meet such involved objectives in context of several applications. In the first part, we focus on autonomous driving; motion planning for autonomous vehicles has to be able to cope with various complex requirements from the rules of the road, presence of (dynamic) obstacles and unusual circumstances, as well as the vehicles’ own kinodynamic and geometric constraints. We specifically target (familiar) situations where all road rules cannot be obeyed simultaneously. We introduce quantitative semantics of Linear Temporal Logic (LTL) and use robustness of Signal Temporal Logic (STL) to recognize maximally-satisfying motion plans and discuss how sampling-based algorithms can be used to compute asymptotically optimal i.e. maximally satisfying motion plans, also in multi-agent settings. In the second part of the talk, we focus on guided exploration of unknown environments; we couple quantitative semantics robustness of STL with RRT * in order to keep an UAV within desired distances from obstacles and points of interest. Finally, we show our recent results on semantic abstraction-guided motion planning that uses learning of semantic dependencies in unknown environments to achieve desired missions more efficiently.

Jana Tumova is an associate professor at the School of Electrical Engineering and Computer Science at KTH Royal Institute of Technology. She received PhD in computer science from Masaryk University and was awarded ACCESS postdoctoral fellowship at KTH in 2013. She was also a visiting researcher at MIT , Boston University, and Singapore-MIT Alliance for Research and Technology. Her research interests include formal methods applied in decision making, motion planning, and control of autonomous systems. Among other projects, she is a recipient of a Swedish Research Council Starting Grant to explore compositional planning for multi-agent systems under temporal logic goals and a WASP Expeditions project focusing on design of socially acceptable and correct-by-design autonomous systems.

This talk is part of the IRLab Seminars: Robotics, Computer Vision & AI series.

This talk is included in these lists:

• IRLab Seminars: Robotics, Computer Vision & AI
• https://bham-ac-uk.zoom.us/j/83482525060?pwd=V2NqMDVKaXVTWWZoR2VTUXA5MmRmUT09

Note that ex-directory lists are not shown.