Refine
Departments, institutes and facilities
Document Type
- Conference Object (51)
- Preprint (10)
- Article (9)
- Part of a Book (3)
- Report (3)
- Research Data (1)
Year of publication
Keywords
- Automatic Short Answer Grading (2)
- Cognitive robot control (2)
- Drosophila (2)
- Explainable robotics (2)
- Learning from experience (2)
- Navigation (2)
- robot execution failures (2)
- robotics (2)
- Active Learning (1)
- Autism Spectrum Disorder (1)
Grasp verification is advantageous for autonomous manipulation robots as they provide the feedback required for higher level planning components about successful task completion. However, a major obstacle in doing grasp verification is sensor selection. In this paper, we propose a vision based grasp verification system using machine vision cameras, with the verification problem formulated as an image classification task. Machine vision cameras consist of a camera and a processing unit capable of on-board deep learning inference. The inference in these low-power hardware are done near the data source, reducing the robot's dependence on a centralized server, leading to reduced latency, and improved reliability. Machine vision cameras provide the deep learning inference capabilities using different neural accelerators. Although, it is not clear from the documentation of these cameras what is the effect of these neural accelerators on performance metrics such as latency and throughput. To systematically benchmark these machine vision cameras, we propose a parameterized model generator that generates end to end models of Convolutional Neural Networks(CNN). Using these generated models we benchmark latency and throughput of two machine vision cameras, JeVois A33 and Sipeed Maix Bit. Our experiments demonstrate that the selected machine vision camera and the deep learning models can robustly verify grasp with 97% per frame accuracy.
Efficient and comprehensive assessment of students knowledge is an imperative task in any learning process. Short answer grading is one of the most successful methods in assessing the knowledge of students. Many supervised learning and deep learning approaches have been used to automate the task of short answer grading in the past. We investigate why assistive grading with active learning would be the next logical step in this task as there is no absolute ground truth answer for any question and the task is very subjective in nature. We present a fast and easy method to harness the power of active learning and natural language processing in assisting the task of grading short answer questions. A webbased GUI is designed and implemented to incorporate an interactive short answer grading system. The experiments show that active learning saves the time and effort of graders in assessment and reaches the performance of supervised learning with less amount of graded answers for training.
Comparative Evaluation of Pretrained Transfer Learning Models on Automatic Short Answer Grading
(2020)
Automatic Short Answer Grading (ASAG) is the process of grading the student answers by computational approaches given a question and the desired answer. Previous works implemented the methods of concept mapping, facet mapping, and some used the conventional word embeddings for extracting semantic features. They extracted multiple features manually to train on the corresponding datasets. We use pretrained embeddings of the transfer learning models, ELMo, BERT, GPT, and GPT-2 to assess their efficiency on this task. We train with a single feature, cosine similarity, extracted from the embeddings of these models. We compare the RMSE scores and correlation measurements of the four models with previous works on Mohler dataset. Our work demonstrates that ELMo outperformed the other three models. We also, briefly describe the four transfer learning models and conclude with the possible causes of poor results of transfer learning models.
This project investigated the viability of using the Microsoft Kinect in order to obtain reliable Red-Green-Blue-Depth (RGBD) information. This explored the usability of the Kinect in a variety of environments as well as its ability to detect different classes of materials and objects. This was facilitated through the implementation of Random Sample and Consensus (RANSAC) based algorithms and highly parallelized workflows in order to provide time sensitive results. We found that the Kinect provides detailed and reliable information in a time sensitive manner. Furthermore, the project results recommend usability and operational parameters for the use of the Kinect as a scientific research tool.
In robot-assisted therapy for individuals with Autism Spectrum Disorder, the workload of therapists during a therapeutic session is increased if they have to control the robot manually. To allow therapists to focus on the interaction with the person instead, the robot should be more autonomous, namely it should be able to interpret the person's state and continuously adapt its actions according to their behaviour. In this paper, we develop a personalised robot behaviour model that can be used in the robot decision-making process during an activity; this behaviour model is trained with the help of a user model that has been learned from real interaction data. We use Q-learning for this task, such that the results demonstrate that the policy requires about 10,000 iterations to converge. We thus investigate policy transfer for improving the convergence speed; we show that this is a feasible solution, but an inappropriate initial policy can lead to a suboptimal final return.
Robots applied in therapeutic scenarios, for instance in the therapy of individuals with Autism Spectrum Disorder, are sometimes used for imitation learning activities in which a person needs to repeat motions by the robot. To simplify the task of incorporating new types of motions that a robot can perform, it is desirable that the robot has the ability to learn motions by observing demonstrations from a human, such as a therapist. In this paper, we investigate an approach for acquiring motions from skeleton observations of a human, which are collected by a robot-centric RGB-D camera. Given a sequence of observations of various joints, the joint positions are mapped to match the configuration of a robot before being executed by a PID position controller. We evaluate the method, in particular the reproduction error, by performing a study with QTrobot in which the robot acquired different upper-body dance moves from multiple participants. The results indicate the method's overall feasibility, but also indicate that the reproduction quality is affected by noise in the skeleton observations.
This paper presents the b-it-bots RoboCup@Work team and its current hardware and functional architecture for the KUKA youBot robot. We describe the underlying software framework and the developed capabilities required for operating in industrial environments including features such as reliable and precise navigation, flexible manipulation, robust object recognition and task planning. New developments include an approach to grasp vertical objects, placement of objects by considering the empty space on a workstation, and the process of porting our code to ROS2.
During robot-assisted therapy, a robot typically needs to be partially or fully controlled by therapists, for instance using a Wizard-of-Oz protocol; this makes therapeutic sessions tedious to conduct, as therapists cannot fully focus on the interaction with the person under therapy. In this work, we develop a learning-based behaviour model that can be used to increase the autonomy of a robot’s decision-making process. We investigate reinforcement learning as a model training technique and compare different reward functions that consider a user’s engagement and activity performance. We also analyse various strategies that aim to make the learning process more tractable, namely i) behaviour model training with a learned user model, ii) policy transfer between user groups, and iii) policy learning from expert feedback. We demonstrate that policy transfer can significantly speed up the policy learning process, although the reward function has an important effect on the actions that a robot can choose. Although the main focus of this paper is the personalisation pipeline itself, we further evaluate the learned behaviour models in a small-scale real-world feasibility study in which six users participated in a sequence learning game with an assistive robot. The results of this study seem to suggest that learning from guidance may result in the most adequate policies in terms of increasing the engagement and game performance of users, but a large-scale user study is needed to verify the validity of that observation.
Improving Robustness of Task Execution Against External Faults Using Simulation Based Approach
(2013)
Robots interacting in complex and cluttered environments may face unexpected situations referred to as external faults which prohibit the successful completion of their tasks. In order to function in a more robust manner, robots need to recognise these faults and learn how to deal with them in the future. We present a simulation-based technique to avoid external faults occurring during execusion releasing actions of a robot. Our technique utilizes simulation to generate a set of labeled examples which are used by a histogram algorithm to compute a safe region. A safe region consists of a set of releasing states of an object that correspond to successful performances of the action. This technique also suggests a general solution to avoid the occurrence of external faults for not only the current, observable object but also for any other object of the same shape but different size.
Robots, which are able to carry out their tasks robustly in real world environments, are not only desirable but necessary if we want them to be more welcome for a wider audience. But very often they may fail to execute their actions successfully because of insufficient information about behaviour of objects used in the actions.
In the field of service robots, dealing with faults is crucial to promote user acceptance. In this context, this work focuses on some specific faults which arise from the interaction of a robot with its real world environment due to insufficient knowledge for action execution.
In our previous work [1], we have shown that such missing knowledge can be obtained through learning by experimentation. The combination of symbolic and geometric models allows us to represent action execution knowledge effectively. However we did not propose a suitable representation of the symbolic model.
In this work we investigate such symbolic representation and evaluate its learning capability. The experimental analysis is performed on four use cases using four different learning paradigms. As a result, the symbolic representation together with the most suitable learning paradigm are identified.
Unexpected Situations in Service Robot Environment: Classification and Reasoning Using Naive Physics
(2014)
In the field of domestic service robots, recovery from faults is crucial to promote user acceptance. In this context we focus in particular on some specific faults, which arise from the interaction of a robot with its real world environment. Even a well-modelled robot may fail to perform its tasks successfully due to unexpected situations, which occur while interacting. These situations occur as deviations of properties of the objects (manipulated by the robot) from their expected values. Hence, they are experienced by the robot as external faults.
In the field of service robots, dealing with faults is crucial to promote user acceptance. In this context, this work focuses on some specific faults which arise from the interaction of a robot with its real world environment due to insufficient knowledge for action execution. In our previous work [1], we have shown that such missing knowledge can be obtained through learning by experimentation. The combination of symbolic and geometric models allows us to represent action execution knowledge effectively. However we did not propose a suitable representation of the symbolic model. In this work we investigate such symbolic representation and evaluate its learning capability. The experimental analysis is performed on four use cases using four different learning paradigms. As a result, the symbolic representation together with the most suitable learning paradigm are identified.
Robot deployment in realistic environments is challenging despite the fact that robots can be quite skilled at a large number of isolated tasks. One reason for this is that robots are rarely equipped with powerful introspection capabilities, which means that they cannot always deal with failures in an acceptable manner; in addition, manual diagnosis is often a tedious task that requires technicians to have a considerable set of robotics skills. In this paper, we discuss our ongoing efforts to address some of these problems. In particular, we (i) present our early efforts at developing a robotic black box and consider some factors that complicate its design, (ii) explain our component and system monitoring concept, and (iii) describe the necessity for remote monitoring and experimentation as well as our initial attempts at performing those. Our preliminary work opens a range of promising directions for making robots more usable and reliable in practice.
Robot deployment in realistic dynamic environments is a challenging problem despite the fact that robots can be quite skilled at a large number of isolated tasks. One reason for this is that robots are rarely equipped with powerful introspection capabilities, which means that they cannot always deal with failures in a reasonable manner; in addition, manual diagnosis is often a tedious task that requires technicians to have a considerable set of robotics skills.