Is this project an undergraduate, graduate, or faculty project?
Graduate
Project Type
group
Campus
Daytona Beach
Authors' Class Standing
Gulsum Tuba Cibuk Girgin, Graduate Student Emre Girgin, Graduate Student
Lead Presenter's Name
Gulsum Tuba Cibuk Girgin
Lead Presenter's College
DB College of Engineering
Faculty Mentor Name
Cagri Kilic
Abstract
Obstacle avoidance is a critical aspect of autonomous mobile robot navigation, particularly in dynamic and static environments. However, traditional avoidance strategies often increase travel time or render traversal impossible, which is problematic in scenarios requiring repeated navigation, such as planetary sample return missions, post-disaster search and rescue, or hazardous environments like nuclear power plants. A more efficient alternative is to manipulate and remove easily displaceable obstacles. Manual intervention, such as astronaut-led operations, is time-consuming, diverts focus from primary tasks, and can be hazardous. Autonomous robotic manipulation offers a solution but introduces challenges in determining whether an obstacle is relocatable based on unknown physical constraints, including object characteristics, surface properties, and interaction dynamics. While object manipulation in controlled environments is well studied, research rarely considers the influence of diverse surface properties, particularly in planetary terrains with spatially varying slopes and soil characteristics. To address this, we propose a probabilistic affordance learning framework that integrates exteroceptive information and proprioceptive feedback to assess whether an obstacle can be removed. By utilizing a robotic arm to interact with obstacles, our approach combines visual data of the object and its supporting surface with force feedback to estimate its relocatability, enabling more effective autonomous navigation in challenging environments.
Did this research project receive funding support (Spark, SURF, Research Abroad, Student Internal Grants, Collaborative, Climbing, or Ignite Grants) from the Office of Undergraduate Research?
No
Learning Rock Pushability on Rough Planetary Terrain
Obstacle avoidance is a critical aspect of autonomous mobile robot navigation, particularly in dynamic and static environments. However, traditional avoidance strategies often increase travel time or render traversal impossible, which is problematic in scenarios requiring repeated navigation, such as planetary sample return missions, post-disaster search and rescue, or hazardous environments like nuclear power plants. A more efficient alternative is to manipulate and remove easily displaceable obstacles. Manual intervention, such as astronaut-led operations, is time-consuming, diverts focus from primary tasks, and can be hazardous. Autonomous robotic manipulation offers a solution but introduces challenges in determining whether an obstacle is relocatable based on unknown physical constraints, including object characteristics, surface properties, and interaction dynamics. While object manipulation in controlled environments is well studied, research rarely considers the influence of diverse surface properties, particularly in planetary terrains with spatially varying slopes and soil characteristics. To address this, we propose a probabilistic affordance learning framework that integrates exteroceptive information and proprioceptive feedback to assess whether an obstacle can be removed. By utilizing a robotic arm to interact with obstacles, our approach combines visual data of the object and its supporting surface with force feedback to estimate its relocatability, enabling more effective autonomous navigation in challenging environments.