Sherman Luo

EECS Department, University of California, Berkeley

Technical Report No. UCB/EECS-2020-44

May 12, 2020

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2020/EECS-2020-44.pdf

What makes interactions challenging for robots that navigate around people in general, and autonomous cars in particular, is the need to account for the mutual influence between the robot's actions and the human's. These interactions are best modeled by dynamic game theory, which in turn motivates the use of game-theoretic planners for autonomous cars. Recent work proposed a number of such planners that leverage trajectory optimization and Model Predictive Control, but adapt them to account for the strategic, multi-agent structure of the problem. In this work, we provide a quantitative and qualitative empirical analysis of the performance of these planners, with the goal of gaining a deeper understanding of their advantages and limitations in challenging interactive driving situations.

Advisors: Anca Dragan

---

BibTeX citation:

@mastersthesis{Luo:EECS-2020-44,
    Author= {Luo, Sherman},
    Title= {Comparing Game Planners},
    School= {EECS Department, University of California, Berkeley},
    Year= {2020},
    Month= {May},
    Url= {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2020/EECS-2020-44.html},
    Number= {UCB/EECS-2020-44},
    Abstract= {What makes interactions challenging for robots that navigate around people in general, and autonomous cars in particular, is the need to account for the mutual influence between the robot's actions and the human's. These interactions are best modeled by dynamic game theory, which in turn motivates the use of game-theoretic planners for autonomous cars. Recent work proposed a number of such planners that leverage trajectory optimization and Model Predictive Control, but adapt them to account for the strategic, multi-agent structure of the problem. In this work, we provide a quantitative and qualitative empirical analysis of the performance of these planners, with the goal of gaining a deeper understanding of their advantages and limitations in challenging interactive driving situations.},
}

EndNote citation:

%0 Thesis
%A Luo, Sherman 
%T Comparing Game Planners
%I EECS Department, University of California, Berkeley
%D 2020
%8 May 12
%@ UCB/EECS-2020-44
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2020/EECS-2020-44.html
%F Luo:EECS-2020-44