VIP projects
Vertically Integrated Projects (VIP) at Notre Dame unite undergraduate education and faculty research in a team-based context. Different from traditional undergraduate research, VIP emphasizes the continuity of the team and the project. First, students working on a VIP team understand that this is a multi-semester/year commitment as no real research question can be solved within a semester. We will form a VIP team based on this project and actively recruit students to our VIP team every semester from different departments/colleges (Electrical, Computer, Mechanical, Physics, Math) as we want to foster a multi-disciplinary working environment. The returning students to the program help to mentor new team members, often across disciplines. In our VIP project, students will participate the development of the algorithms and software, and will participate the experiments and data collection.
The whole group – VIP students, PI and PI’s graduate students – will meet weekly to discuss the project. Participation in VIP will provide course credits that will be counted towards the students’ senior design requirements. VIP will provide a unique opportunity for undergraduate students to explore and develop comprehensive applications of engineering technologies that they cannot learn from traditional courses.
1. Integrated Task and Motion Planning for Robotic Systems
2. Human-Robot Collaboration Team
Goals:
Student's Role:
Methods:
In Human-Robot Collaboration, robots are expected to work next to human in warehouses, daily housekeeping, and other robot assistant applications safely, intelligently and friendly. To achieve this goal, the robotic system should be equipped with capacities of understanding intentions of human partners and reasoning according to the behaviors of human partners and the state of the environment. The main idea is to combine the learning-based approach with traditional high-level task planning algorithms.
Develop algorithms to track human movements and collect data from demonstrations. Build human models using Bayesian Non-parametric learning algorithms to ensure robots could infer the human intention correctly. Develop high-level task planning algorithms to enable the robot to behave collaboratively with human partners. Implement these algorithms on the Baxter robot.
Bayesian Non-parametric learning, Learning from demonstrations, Formal Methods, Dynamic Programming, Stochastic
Contact Information: Contact