Publications

@inproceedings{10.1007/978-3-031-15908-4_5,

title = {Trigger-Assisted Ambidextrous Control Framework for Teleoperation of Two Legged Manipulators},

author = {Christopher Peers and Joseph Humphreys and Yuhui Wan and Jun Li and Jingcheng Sun and Robert Richardson and Chengxu Zhou},

editor = {Salvador Pacheco-Gutierrez and Alice Cryer and Ipek Caliskanelli and Harun Tugal and Robert Skilton},

isbn = {978-3-031-15908-4},

year  = {2022},

date = {2022-03-01},

urldate = {2022-03-01},

booktitle = {Towards Autonomous Robotic Systems},

pages = {50--62},

publisher = {Springer International Publishing},

address = {Cham},

abstract = {This paper presents a motion-capture based control framework for the purpose of effectively teleoperating two legged manipulators without significant delays caused by the switching of controllers. The control framework generates high-level trajectories in 6 degrees of freedom and uses finger gesture detection to act as triggers in selecting which robot to control as well as toggling various aspects of control such as yaw rotation of the quadruped platform. The functionality and ease of use of the control framework are demonstrated through a real-life experiment where the operator controls two quadrupedal manipulator robots to open a spray can. The experiment was successfully accomplished by the proposed teleoperation framework.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{10.1007/978-3-031-15908-4_6,

title = {Teleoperating a Legged Manipulator Through Whole-Body Control},

author = {Joseph Humphreys and Christopher Peers and Jun Li and Yuhui Wan and Jingcheng Sun and Robert Richardson and Chengxu Zhou},

editor = {Salvador Pacheco-Gutierrez and Alice Cryer and Ipek Caliskanelli and Harun Tugal and Robert Skilton},

isbn = {978-3-031-15908-4},

year  = {2022},

date = {2022-03-01},

booktitle = {Towards Autonomous Robotic Systems},

pages = {63--77},

publisher = {Springer International Publishing},

address = {Cham},

abstract = {In this work, we present a highly functional teleoperation system, that integrates a full-body inertia-based motion capture suit and three intuitive teleoperation strategies with a Whole-Body Control (WBC) framework, for quadrupedal legged manipulators. This enables the realisation of commands from the teleoperator that would otherwise not be possible, as the framework is able to utilise DoF redundancy to meet several objectives simultaneously, such as locking the gripper frame in position while the trunk completes a task. This is achieved through the WBC framework featuring a defined optimisation problem that solves a range of Cartesian and joint space tasks, while subject to a set of constraints (e.g. halt constraints). These tasks and constraints are highly modular and can be configured dynamically, allowing the teleoperator to switch between teleoperation strategies seamlessly. The overall system has been tested and validated through a physics-based simulation and a hardware test, demonstrating all functionality of the system, which in turn has been used to evaluate its effectiveness.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{10.1145/3546932.3547014,

title = {A Benchmark for Active Learning of Variability-Intensive Systems},

author = {Shaghayegh Tavassoli and Carlos Diego N. Damasceno and Mohammad Reza Mousavi and Ramtin Khosravi},

url = {https://doi.org/10.1145/3546932.3547014},

doi = {10.1145/3546932.3547014},

isbn = {9781450394437},

year  = {2022},

date = {2022-03-01},

booktitle = {Proceedings of the 26th ACM International Systems and Software Product Line Conference - Volume A},

pages = {245–249},

publisher = {Association for Computing Machinery},

address = {Graz, Austria},

series = {SPLC '22},

abstract = {Behavioral models are the key enablers for behavioral analysis of Software Product Lines (SPL), including testing and model checking. Active model learning comes to the rescue when family behavioral models are non-existent or outdated. A key challenge on active model learning is to detect commonalities and variability efficiently and combine them into concise family models. Benchmarks and their associated metrics will play a key role in shaping the research agenda in this promising field and provide an effective means for comparing and identifying relative strengths and weaknesses in the forthcoming techniques. In this challenge, we seek benchmarks to evaluate the efficiency (e.g., learning time and memory footprint) and effectiveness (e.g., conciseness and accuracy of family models) of active model learning methods in the software product line context. These benchmark sets must contain the structural and behavioral variability models of at least one SPL. Each SPL in a benchmark must contain products that requires more than one round of model learning with respect to the basic active learning L* algorithm. Alternatively, tools supporting the synthesis of artificial benchmark models are also welcome.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}