Skip to content

Second Seminar in 2023-2024 Seminar Series

We're excited to continue our virtual seminar series for the TC on Model-Based Optimization for Robotics.

We'll have a talk from Alessandro Saccon from TU/e at 9AM EST January 31st 2024 (Wednesday). Please find the flyer with all the details at this link.

Speaker: Alessandro Saccon (Eindhoven University of Technology (TU/e))

Title: Impact-Aware Manipulation: Exploiting Intentional Collisions to Speed Up Task Execution

Abstract: Traditionally, robots fear collisions. Contact is either avoided at all cost or established at vanishing speed. There are several reasons for this to happen: an important one is the lack of suitable hardware. Robot physical components, especially joints and transmissions, need to be made resilient to collisions by design in order not to get damaged. However, recent hardware developments, including those in the area of soft robotics, are rapidly changing the hardware landscape. Modern robot hardware is starting to showcase the possibility of sustaining and thus exploiting impacts. A clear and well-known example is Boston Dynamics' humanoid robot ATLAS, performing acrobatic moves such as somersaults and jumps. The availability of impact-resilient hardware opens up possibilities to develop and validate new impact-aware manipulation and locomotion strategies. Strategies where contact between the robot and irs surrounding environment is intentionally established at desired non-negligible speed to achieve user-defined motion objectives. The exploitation of intentional impacts is particularly appealing for reducing a task's execution time. But also for reducing energy consumption due to the lack of fast decelerations and acceleration traditionally required to stop the robot at contact time and to make it move again after contact has been established. In this talk, I will provide an overview of the research carried out at the TU/e on this fascinating new field of research, specifically focusing on impact-aware control and post-impact velocity prediction aspects.

Date: Wednesday, January 31st 2024

Time: 09:00-10:00 AM EST (GMT -05:00)

Link: https://columbiauniversity.zoom.us/j/91247893326?pwd=L2JWU21aQzc4cU1ZQklEb0QrWGQvdz09

More details on upcoming seminars (and video links for past ones) can be found here.