| 1. |
- Barbosa, Fernando S., 1992-, et al.
(författare)
-
Formal Methods for Robot Motion Planning with Time and Space Constraints (Extended Abstract)
- 2021
-
Ingår i: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). - Cham : Springer Nature. ; , s. 1-14
-
Konferensbidrag (refereegranskat)abstract
- Motion planning is one of the core problems in a wide range of robotic applications. We discuss the use of temporal logics to include complex objectives, constraints, and preferences in motion planning algorithms and focus on three topics: the first one addresses computational tractability of Linear Temporal Logic (LTL) motion planning in systems with uncertain non-holonomic dynamics, i.e. systems whose ability to move in space is constrained. We introduce feedback motion primitives and heuristics to guide motion planning and demonstrate its use on a rover in 2D and a fixed-wing drone in 3D. Second, we introduce combined motion planning and hybrid feedback control design in order to find and follow trajectories under Metric Interval Temporal Logic (MITL) specifications. Our solution creates a path to be tracked, a sequence of obstacle-free polytopes and time stamps, and a controller that tracks the path while staying in the polytopes. Third, we focus on motion planning with spatio-temporal preferences expressed in a fragment of Signal Temporal Logic (STL). We introduce a cost function for a of a path reflecting the satisfaction/violation of the preferences based on the notion of STL spatial and temporal robustness. We integrate the cost into anytime asymptotically optimal motion planning algorithm RRT ⋆ and we show the use of the algorithm in integration with an autonomous exploration planner on a UAV.
|
|
| 2. |
- Barbosa, Fernando S., 1992-, et al.
(författare)
-
Risk-Aware Motion Planning in Partially Known Environments
- 2021
-
Ingår i: 2021 60th IEEE conference on decision and control (CDC). - : Institute of Electrical and Electronics Engineers (IEEE). ; , s. 5220-5226
-
Konferensbidrag (refereegranskat)abstract
- Recent trends envisage robots being deployed inareas deemed dangerous to humans, such as buildings with gasand radiation leaks. In such situations, the model of the underlying hazardous process might be unknown to the agent a priori, giving rise to the problem of planning for safe behaviour inpartially known environments. We employ Gaussian Process regression to create a probabilistic model of the hazardous process from local noisy samples. The result of this regression is then used by a risk metric, such as the Conditional Value-at-Risk, to reason about the safety at a certain state. The outcome is a risk function that can be employed in optimal motion planning problems. We demonstrate the use of the proposed function in two approaches. First is a sampling-based motion planning algorithm with an event-based trigger for online replanning. Second is an adaptation to the incremental Gaussian Process motion planner (iGPMP2), allowing it to quickly react and adapt to the environment. Both algorithms are evaluated in representative simulation scenarios, where they demonstrate the ability of avoiding high-risk areas.
|
|
| 3. |
- Grover, Kush, et al.
(författare)
-
Semantic Abstraction-Guided Motion Planningfor scLTL Missions in Unknown Environments
- 2021
-
Ingår i: Robotics: Science and Systems. - : RSS FOUNDATION-ROBOTICS SCIENCE & SYSTEMS FOUNDATION.
-
Konferensbidrag (refereegranskat)abstract
- Complex mission specifications can be often specifiedthrough temporal logics, such as Linear Temporal Logic and itssyntactically co-safe fragment, scLTL. Finding trajectories thatsatisfy such specifications becomes hard if the robot is to fulfilthe mission in an initially unknown environment, where neitherlocations of regions or objects of interest in the environmentnor the obstacle space are known a priori. We propose an algorithmthat, while exploring the environment, learns importantsemantic dependencies in the form of a semantic abstraction,and uses it to bias the growth of an Rapidly-exploring randomgraph towards faster mission completion. Our approach leadsto finding trajectories that are much shorter than those foundby the sequential approach, which first explores and then plans.Simulations comparing our solution to the sequential approach,carried out in 100 randomized office-like environments, showmore than 50% reduction in the trajectory length.
|
|
