Trajectory Optimization among Decision-Making Agents: Self-Driving Cars, Drones and Multi-Robot SystemsWe move towards an era of smart cities, where autonomous cars will provide on-demand transportation while making our streets safer and mobile robots will coexist with humans. For urban driving the motion plan of the autonomous car must account for the interaction with the other traffic participants and consider that they are as well decision-making entities. For example, when humans drive a car they are fully aware of their environment and how other drivers and pedestrians may react to their actions. In the first part of this talk I will give an overview of our work in motion planning among other decision-making vehicles. I will first introduce a receding-horizon trajectory optimization framework, which we have employed for parallel autonomy and for autonomous driving. Then I will introduce two methods for coordination with other vehicles: via explicit communication and distributed optimization, or via implicit coordination based on a joint behavior estimation and trajectory optimization framework, where we can model the multimodal uncertainty in the predictions. In the second part of this talk I will cover our recent work on multi-robot coordination, including formation control, chance-constrained collision avoidance and learning of communication policies, all applied to teams of micro-aerial vehicles.
Keynote Speeches *Ranked by speech time Keynote Speech 1
2020-11-27
173601 byte
Conference paper
Electronic Resource
English
IEEE | 2020
Keynote Speech: Evolution of Automobiles
British Library Conference Proceedings | 1996
|Keynote Speech: An Uncertain World
British Library Conference Proceedings | 2005
|Architectural Association Mars Symposium Keynote Speech
NTIS | 2021
|