Publications

• Sochaki, G., Iida, F. and Hughes, J. (2021) ‘Compliant Sensorized Testing Device to Provide a Model-Based Estimation of the Cooking Time of Vegetables‘, 16th International Conference on Intelligent Autonomous Systems.
• Sochacki, G., Hughes, J., Hauser, S., and Iida, F. (2021) ‘Closed-Loop Robotic Cooking of Scrambled Eggs with a Salinity-based ‘Taste’ Sensor‘, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).
• Sochacki, G., Hughes, J., and Iida, F. (2022) ‘Sensorized Compliant Robot Gripper for Estimating the Cooking Time of Boil-Cooked Vegetables‘, Intelligent Autonomous Systems 16 (IAS 2021).
• Sochacki, G., Abdulali, A., and Iida, F. (2022) ‘Mastication-Enhanced Taste-Based Classification of Multi-Ingredient Dishes for Robotic Cooking‘, Frontiers in Robotics and AI-Bio-Inspired Robotics.
• Sochacki, G., Abdulali, A., Cheke, L., and Iida, F.(2022) Theoretical Framework for Human-Like Robotic Taste with Reference to Nutritional Needs. IOP Conference Series: Materials Science and Engineering, 1292.
  
• Shi, J., Abdulali, A., Sochacki, G., & Iida, F. (2023) Closed-Loop Robotic Cooking of Soups with Multi-modal Taste Feedback. Towards Autonomous Robotic Systems. TAROS 2023: Lecture Notes in Computer Science, 14136. Springer, Cham
• Sochacki, G., Abdulali, A., Khadem Hosseini, N., & Iida, F. (2023) Recognition of Human Chef’s Intentions for Incremental Learning of Cookbook by Robotic Salad Chef. IEEE Access, vol. 11, pp. 57006-57020, 2023.

Presentations

• “Residual Physics for Grasp Failure Prediction” (oral) – Lincoln Conference on Intelligent Robots and Systems 2020 [October 2020] – Online.
• “Touch, smell, taste and sound sensing in robotic kitchen” (oral) – International Workshop on Embodied Intelligence 2021 [March 2021] – Online.
• “Visual Serving for Human Tracking and Counting” (oral) – AgriFoRwArdS CDT Summer School 2021 [June 2021] – Online.
• “Compliant Sensorized Testing Device to Provide a Model Based Estimation of the Cooking Time of Vegetables” (oral) – International Conference on Intelligent Autonomous Systems (IAS) 2021 [June 2021] – Online.
• “Closed Loop Action for Robotic Chef Implementation” (oral) – AgriFoRwArdS CDT Annual Conference 2021 [July 2021] – Online.
• “Visual Serving for Human Tracking and Counting” (oral) – AgriFoRwArdS CDT Annual Conference 2021 [July 2021] – Online.
• “Closed-Loop Robotic Cooking of Scrambled Eggs with a Salinity-based ‘Taste’ Sensor” (oral) – International Conference on Intelligent Robots and Systems (IROS) 2021 [October 2021] – Online.
• “Theoretical Framework for Human-Like Robotic Taste with Reference to Nutritional Needs” (oral) – International Conference on Embodied Intelligence 2022 [March 2022]- Cambridge, UK.
• “Unknown title” (oral) – Robosoft 2022 [April 2022] – Edinburgh, UK.
• “Sensorized Compliant Robot Gripper for Estimating the Cooking Time of Boil-Cooked Vegetables” (oral) – International Conference on Intelligent Autonomous Systems (IROS) 2022 [June 2022] – Online.
• “Implementation of taste-enabled robotic chef” (oral) – AgriFoRwArdS CDT Annual Conference 2022 [June 2022] – Lincoln, UK.
• “Stand-Alone, Easy-to-Scale and Low-Overhead Robotic Fish & Chips Dark Kitchen” (oral) – AgriFoRwArdS CDT Summer School 2022 [July 2022] – Norwich, UK.
• “Does Baxter Dream of Electric Beans?” (oral) – AgriFoRwArdS CDT Summer School 2023 [March 2023] – Lincoln, UK.
• “Closed-Loop Robotic Cooking of Soups with Multi-modal Taste Feedback” (oral) – Towards Autonomous Robotic Systems (TAROS) 2023 / AgriFoRwArdS CDT Annual Conference 2023 / Joint Robotics CDT Annual Conference 2023 [September 2023] – Cambridge, UK.
• “Improving Robotic Taste Performance in Classification Task with Mastication” (poster) – Towards Autonomous Robotic Systems (TAROS) 2023 / AgriFoRwArdS CDT Annual Conference 2023 / Joint Robotics CDT Annual Conference 2023 [September 2023] – Cambridge, UK.

Other Activities

• Co-Chaired a breakout session at the International Workshop on Embedded Intelligence 2021
• Designed the robotic kits used for the AgriFoRwArdS Summer School 2021, and provided significant technical support for the event
• Guest on BBC Radio Berkshire
• Discussion Panel member at the AgriFoRwArdS CDT Annual Conference 2021 – Discussion topic: Tony Pridmore’s keynote talk “Plant Phenotyping: Getting to the root of the problem” [July 2021] – Online.
• UKRAS Robot Lab Live 2022 demonstration
• Entered the PUB.R Competition at the International Conference on Robotics and Automation (ICRA) 2023.

MSc Project

Residual Physics for Grasp Failure Prediction

Prediction of grasping success is not a solved problem, with current research focusing on the grasp stability during lifting an object, which is much less then human intuition can do. Human intuition can assess the extent of  possible movements, that can be done without losing a grasp of an object. The project attempts to produce an algorithm, which can analyze trajectory plans for a robotic arm and decide if the grasp would remain stable, based on tactile information, series of waypoints, and estimates of object mass and inertia. The chosen approach is to use residual physics, where a coarse physical model is complemented by residua l computed by a neural network. The project hopes to enable choosing optimal paths and maximum speeds for not optimal grasps.

PhD Project

Taste-Enabled Robotic Chef – On Robots Learning to Cook from Taste Feedback and Human Demonstration

Cooking and consuming food is an important part of human society and culture. Regardless of technological advances, food preparation is still a time-consuming chore most people do daily. Cooking could be automated by introducing robotic chefs, which are robots capable of cooking a significant selection of dishes. This project focuses on exploring how hardware, both actuating and sensing, works in conjunction with control and machine learning algorithms to form a feedback loop in the context of cooking. Robotic chef faces many challenges including sensing properties of food, manipulation and learning from a limited amount of data, but the biggest challenge is the subjective nature of assessing the outcome of cooking. This problem is inescapable as the final dish is judged by the diner who is inherently subjective and the same dish may have a very different palatability for different diners.

This project contributes to research in sensing and learning of the state and palatability of a dish cooked by a robot. It includes using tactile sensing in a robot that presented a raw and well-cooked vegetable to assess readiness and predict the course of further cooking. The project also discusses the use of electronic taste as feedback in the cooking process, where the robot replicates a variation of a dish preferred by a human diner. It was also proven that replication of the chewing process improves electronic taste and allows better classification between variations of dishes. The use of cameras to program robotic chefs by visual demonstration is also elaborated. Novel methods of machine learning for food palatability assessment are also discussed. Finally, most of the methods and systems presented have some subjective input from a human that allows the robot to deal with the subjectivity of food taste by catering to this specific person.

In summary, the project presents significant progress in research into robotic chefs, contributing to all parts of robotic chefs including manipulation, sensing, signal processing and learning. Moreover, it is the first work that tackles robotic cooking with the use of electronic taste and catering to the specific and subjective preferences of a human diner.

Greg’s PhD project was completed in collaboration with Beko, and with primary supervision by Dr Fumiya Iida.