FlexCRAFT

The FlexCRAFT program

FlexCRAFT is a Dutch research program to advance Cognitive Robotics for Flexible Agro-Food Technology. The program is executed by an excellent team of robotic researchers from five universities in collaboration with fifteen companies and supported by NWO/TTW.

Projects

The program consists of seven projects organized in two research lines (RL). In RL1, generic robotic capabilities are developed to allow robots to deal with the variations present in agro-food environments. In RL2, these generic capabilities are integrated, tested and evaluated in three agro-food use-cases. RL1 – Development of generic robotic capabilities P1: Active Perception…

Four generic research projects

To deal with these challenges, generic robotic methodologies will be developed in the four generic research projects, P1, P2, P3 and P4. These generic methodologies will be integrated, tested and evaluated in the use-case demonstrators.

FlexCRAFT

Four generic projects

Three use-case projects

The program revolves around three agro-food use-cases. The uses-cases give rise to generic robotic challenges in dealing with variation. Demonstrators for each of the use-cases will be developed in the use-case projects P5, P6 and P7.

Wageningen University

Farm Technology Group Prof.dr.ir. Eldert van Henten, program leader. Eldert is the head of the Farm Technolo- gy Group. He will contribute with his vast knowledge of agro-food, 20 years’ experience with vision and robotics in agriculture, and robotic systems integration resulting in proof-of-concept systems of harvesting robots, vi- sion-based weeding systems, egg-collecting robots, and…

University of Eindhoven

Control Systems Technology Group Prof.dr.ir. Maarten Steinbuch. Maarten is a Distinguished University Professor and head of the Control Systems Technology Group within the Mechanical Engineering Department. He will contrib- ute with his vast experience with (control of) mechatronic and robotic systems. Prof.dr.ir. Herman Bruyninckx, project leader of P7 “Packaging Robotics”. Herman is a part-time professor…

University of Twente

Precision Engineering Prof.dr.ir. Dannis Brouwer PDEng, project leader of P4. Dannis is an associate professor (with promotion rights) holding the chair of Precision Engineering. He will contribute with his substantial background in the high-tech systems industry (3 year at Philips CFT and 5 years at Demcon) and his knowledge of flexures for large range of…

Delft University of Technology

Cognitive Robotics Prof.dr.ir. Robert Babuska, research line leader of RL1. Robert is a full professor of Intelligent Con- trol and Robotics and section leader of Learning and Autonomous Control in the Cognitive Robotics Group and will contribute with over 25 years of experience with research in computational intelligence, machine learning, control, and system identification, including…

University of Amsterdam

Informatics Institute Prof.dr. Theo Gevers, project leader of P1. Theo is a full professor and head of the Computer Vision group at the University of Amsterdam. He will contribute with his vast knowledge of computer vision, perception, machine learning, deep learning, 3D (object) reconstruction, and image understanding, with appli- cations in the agro-food, healthcare, and…

P1: Active Perception

Most of the current robotic systems are based on a traditional sense-plan-act cycle, where perception and action are viewed as individual processes. This approach, however, cannot deal with the challenges that the agro-food environment poses. As perception and action are tightly coupled through the interactions of the robot with the environment, the scientific challenge is…

P2: World Modeling

Current robotic systems are pre-programmed to deal with very specific tasks on a limited set of objects very well-defined in terms of location, shape, size and material properties. To deal with variability and enable flexibility, robotic systems need to reason about the objects in their environment, or world. To that end, they need (to build)…

P3: Planning and Control

In mainstream robotics, task plans are almost exclusively imperative, that is, they consist of reci- pes of motion, sensing and decision-making actions to be realized by the robot. Those recipes are created at development time and are therefore inflexible and unable to deal with variations in objects and environments. The scientific challenge is develop declarative…

P4: Gripping and Manipulation

Current robotic gripping technology is able to handle well-defined, rigid objects. However, grippers for harvesting, food processing, and packaging have to perform their tasks under demanding requirements, such as robustness to variability in product size, shape, and softness, and fast operation to reduce cycle times. State-of-the-art robotic technology does not address these challenges. This large…

P5: Greenhouse Robotics

The greenhouse robotics use case will address the removal of leaves (deleafing) and ripe fruits (harvesting) from tomato plants, two important plant-maintenance operations. Using the capabilities developed in projects P1 to P4 this use-case addresses the challenge of how to deal with variations in the environment, such as changes in illumination and humidity, as well…