Research Projects

Research in our group can be roughly divided into the following areas:

  1. Autonomy, with a special focus on self-driving cars;
  2. Mobility and transportation systems;
  3. Theoretical foundations in systems, perception and control. 

A brief description of these areas with some of the ongoing research projects follows.  

 

Autonomy and self-driving cars

The goal of this research effort is to enable vehicles such as cars and airplanes to safely and reliably drive/fly themselves in an uncertain, dynamic world (public roads, national airspace).

Gokart platform

Part of the research is carried out on our go-kart platform which allows to benchmark algorithms beyond the common safety measures imposed by public roads. See additional info here.

Co-design of Embodied Intelligence 

Designing embodied intelligence systems requires integrating diverse hardware and software components while avoiding excessive complexity. The goal is to ensure safety and efficiency while minimizing resources such as cost, energy, computational demands, and weight, all while considering system interactions. Our research focuses on addressing these integration challenges and strategies, emphasizing task-driven, optimal hardware and software selection to achieve a balance between safety, efficiency, and resource conservation.

Please contact for more information.

 




Illustration of the co-design problem for embodied intelligence, specifically applied to autonomous vehicles. (Visualization by ETH Zürich / Dejan Milojevic)

Motion planning and decision making


Real-time algorithms for control in a complex environment, shared with human and robotic actors

AV at a busy intersection

Regulatory compliance, verification and validation


Methods and tools to design provably correct control systems.

Trajectories evolve both in the physical space and in the logical space defined by rules

High-performance control 


Control at the boundaries of the operational envelope – emergency handling, racing, acrobatics

Autonomous doughnuts

Mobility and Transportation Systems

The goal of this research effort is to develop advanced control and optimization methods to enable new concepts for large-scale transportation systems.

 

Mobility on demand

Manage fleets of 100,000+ autonomous vehicles to provide on-demand mobility, trading off quality of service and operational costs.

load balancing diagram for mobility on demand

 

Robustness and efficiency in large-scale systems


Design distributed control systems that provably maximize throughput, minimize delays, and increase resilience of transportation networks.  

Backpressure algorithm for distributed traffic signal control

Theoretical foundations

There is nothing as practical as a good theory.

We support our applied research projects above by advancing the state of the art in foundational areas such as

  • Control theory and Game theory
  • Algorithmic robotics
  • System design, optimization, and co-design
  • Learning theory
JavaScript has been disabled in your browser