Courses

This course is intended to be a computational introduction to modeling, analysis,  and control of autonomous systems for enhanced physical interaction (mobility and manipulation) with the world. 

• The first part of the course deals with the theoretical frameworks for modeling, analysis (kinematics and dynamics) and control of generic articulated robotic systems, rooted in rich traditions of mechanics and geometry. 

• The rest of the course examines harnessing and enhancing the mobility and manipulation performance of articulated multi‑body mechanical systems in the context of serial-­ and parallel-chain manipulators, as well as wheeled mobile robots (and hybrid combinations of these systems). 

• Case studies are used to highlight this process of systematic performance (mobility and manipulation) evaluation and enhancement for exemplary multi-body robotic systems. 

This course is intended to survey the state-of-the-art in the rapidly evolving field of varying-grades of autonomy for on-road and off-road ground-vehicles. This course will introduce students to both the fundamental advances in science as well as technology behind the systems in a number of application arenas. Topical coverage includes:

• Wheeled-locomotion-platform architectures, from unicycle to multiwheel, and their role in enabling autonomy;

• Sensing and methodologies for merging varying spatial and temporal fidelity sensing to provide situational-awareness;

• Planning frameworks including obstacle-avoidance; waypoint navigation; reactive-behaviors with a focus on merger of behaviors; computational complexity and real-time challenges;

• Control architectures from adaptive cruise control, lane maintenance-assist, self-driving cars.

• Computer-aided engineering analysis tools to support design, analysis, development and validation

Autonomous vehicles are emerging as integral components of the US economy, and there is a critical need for students trained in the theories as well as the technologies surrounding such systems. The development and deployment of scaled (1/10th and 1/5th) autonomous remote-control (RC) vehicle infrastructure offers an affordable, customizable yet integrative focus to support the necessary education and training activities in Safe, Secure, Coordinated and Efficient Autonomy.

The course adopts a project-based learning approach, where students will work on projects within a larger integrative theme of scaled autonomous RC-vehicles. Projects (to be pursued either individually or in 2-3 person teams) will focus on the development and validation of existing cutting-edge approaches to autonomous perception, planning, control and coordination.

Examples of projects could include:

• Comparative analysis of SLAM techniques for improved situational-awareness

• Comparative analysis of vision-based methods for autonomous RC-vehicle control

• Comparative evaluation of autonomous behaviors with vehicle-dynamic simulation (ADAMS/V-REP/Pre-Scan)

• Comparative evaluation of extent/role of visualization in teleoperated control of simulated/actual autonomous RC cars

• Comparative evaluation of extent/role of varying grades of mediated teleoperated control of simulated/actual autonomous RC cars)

This course discusses fundamental concepts in the dynamic behavior of ground vehicles, mainly two and four-wheeled vehicles. It stresses the application of dynamic systems modeling and analysis to understand ride performance, handling, and straight-line running. Practical considerations in vehicle design and its influence on vehicle performance are discussed.

Pre-req: ME 3050 or equivalent or permission of instructor