Motion planning plays a key role in the autonomous control of mobile robots. With the advancement of technology and the introduction of increasingly sophisticated autonomous vehicles, such as self-driving cars, delivery drones, and warehouse robots, complex motion planning objectives and constraints need to be considered. The objective for an autonomous car could be the minimisation of fuel cost or travel time for instance. Possible constraints for a drone include the requirement of avoiding no-fly zones and respecting speed limits. Optimisation-based planning schemes are well suited to address such problems. They also offer the benefit of being able to address requirements on robustness to environmental uncertainty.
When a robot has a more complicated mission than point-to-position navigation from its starting point to the desired destination, it becomes challenging to formulate the path planning problem appropriately. Temporal Logic specifications define acceptable sequences of the state of the robot, where simple tasks (e.g., reaching or avoiding certain areas) are combined with logic (e.g. visit Area A OR Area B) and temporal conditions (e.g. stay safe in Area C UNTIL going to Area D) to capture complex navigation objectives. These specifications can be incorporated into optimisation-based motion planning as mixed-integer constraints.
In this project, robot navigation problems will be formulated as mixed-integer programs. Temporal logic specifications will be derived and encoded to capture specific complex robot objectives. Different formulations will be considered and evaluated based on performance, complexity, and robustness to uncertainty. The second focus of this project will be on exploring methods to solve the resulting mixed-integer programs efficiently. In particular, recently developed approaches that exploit structure by leveraging techniques used in machine learning will be investigated with the objective of solving the mixed-integer programs online. This project will mainly consist of theoretical work but also offers the potential to apply and test the derived results on real mobile robots.
Requirements and key learnings:
This project is ideal for a student who is interested in conducting theoretical research and applying cutting-edge methods. The student should have good background knowledge of optimisation theory and will learn about mixed-integer programming and temporal-logic planning.
This project can also be extended into a master project (PDM).
Background reading:
->https://ieeexplore.ieee.org/abstract/document/7039363,
->https://arxiv.org/abs/1907.02206.
Interested students should contact Tony Wood.