DISTRIBUTED SENSING, CONTROL AND DECISION MAKING IN MULTIAGENT AUTONOMOUS SYSTEMS
Lead Research Organisation:
University of Sheffield
Department Name: Automatic Control and Systems Eng
Abstract
Autonomous intelligent systems will find important applications in our future society. Initial applications will be in the following areas: surveillance, intelligence gathering and operational control in the areas of disaster mitigation (earthquake, nuclear catastrophe, military combat, oil-spills at sea, transport infrastructure breakdown, analysis and assistance with terrorist attacks), space exploration at remote locations (at Trojan asteroids, on Mars and in orbit observations around planets, deep underwater explorations and robotics for offshore oil exploration disasters) followed by large scale applications such as agricultural, search and rescue, manufacturing, and autonomous household robots. These autonomous system will require quick, appropriate, and at the same time informative-to-partners, actions by teams of robots. They can also be computing network based intelligent agents with sensing and control capabilities. It will be a societal requirement that these (semi-)autonomously operating systems to inform their human supervisors about the reasoning behind their actions and their future plans in concise notes for their safety and acceptability by society.
Network based software agents have been in use by our society for some time. Our society is going through information exchange revolution that is developing towards networked intelligent devices. Many of these infrastructure systems are based on well defined discrete inputs and outputs either from human operators or from low dimensional sensor measurements. Little progress has however been made in robot intelligence of autonomy where high complexity, changing environment is to be sensed, reasoned about and acted upon quickly. Partial results have been reported in DARPA, Robocup projects that do not provide comprehensive systematic approach or are not fully publicly available. Progress has only been made in heavily infrastructured environments of robots.
We do not yet have the methodology for a set of autonomous vehicles or agent systems to operate reliably and (semi-)autonomously in complex infrastructure-free environments to solve problems efficiently with minimal human supervision. The reason is that current intelligent agent technology does not provide solutions. Sensor networks with simple computational nodes, that were developed for low power and computational resources do not provide solutions. They miss the ability of high complexity conceptual abstractions onboard a single agent. The computations of these type of agents cannot be substituted by data fusion of low complexity agents due to typical real-time and communication bottlenecks. Methods of multi-agent decentralized decision theory have been developed and very successfully used prior to this project but have not been properly exploited for multiple complex agents.
This project intends to develop a new methodology for autonomous cooperating multi-agent systems that is to boost the technological capabilities of our partner companies and the robotics industry in general. The project will provide the missing capabilities of abstractions concerning world modeling, situational awareness, learning and information management onboard a single agent. These capabilities will enable efficient realtime decision making within multi-agent cooperation and decentralized decision making in poorly structured or infrastructure free environments. These methods will connect digital computing power with human conceptual structures to enable robots to model the world with layers of high and low level concepts as humans do.
Network based software agents have been in use by our society for some time. Our society is going through information exchange revolution that is developing towards networked intelligent devices. Many of these infrastructure systems are based on well defined discrete inputs and outputs either from human operators or from low dimensional sensor measurements. Little progress has however been made in robot intelligence of autonomy where high complexity, changing environment is to be sensed, reasoned about and acted upon quickly. Partial results have been reported in DARPA, Robocup projects that do not provide comprehensive systematic approach or are not fully publicly available. Progress has only been made in heavily infrastructured environments of robots.
We do not yet have the methodology for a set of autonomous vehicles or agent systems to operate reliably and (semi-)autonomously in complex infrastructure-free environments to solve problems efficiently with minimal human supervision. The reason is that current intelligent agent technology does not provide solutions. Sensor networks with simple computational nodes, that were developed for low power and computational resources do not provide solutions. They miss the ability of high complexity conceptual abstractions onboard a single agent. The computations of these type of agents cannot be substituted by data fusion of low complexity agents due to typical real-time and communication bottlenecks. Methods of multi-agent decentralized decision theory have been developed and very successfully used prior to this project but have not been properly exploited for multiple complex agents.
This project intends to develop a new methodology for autonomous cooperating multi-agent systems that is to boost the technological capabilities of our partner companies and the robotics industry in general. The project will provide the missing capabilities of abstractions concerning world modeling, situational awareness, learning and information management onboard a single agent. These capabilities will enable efficient realtime decision making within multi-agent cooperation and decentralized decision making in poorly structured or infrastructure free environments. These methods will connect digital computing power with human conceptual structures to enable robots to model the world with layers of high and low level concepts as humans do.
Planned Impact
The high level of autonomy levels achieved (up to 4b on the extended PACT scale and levels 7-9 on the AFRL scale by Clough) will include decentralised decision making and control by a team of autonomous intelligent agents, including coalition formation for specific goals. These results will have a major impact on industrial automation in general.
The demonstrations will demonstrate to the public the high relevance of this research to society and efficient spending of research funds in terms of the multi-agent vehicle systems. The relevance of this research for networked intelligent autonomous agents with sensing and actuator capabilities for fast response in infrastructure breakdown mitigation in rail transport, nuclear and complex manufacturing systems, will also be demonstrated.
The academic and industrial novelty of the ability of the agents to read system English (sEnglish) technical documents will raise international interest as our agents will read about new physical and mental skills, such as movement controls, movement planning, procedures of payload tasks, situational awareness techniques, response rules and behaviour constraint handling.
Our methods of building up layered abstractions of agent operations, for their physical self-control, communications and team working capabilities, will be academically significant and will show the way for future software engineering in the area of autonomous systems.
The above impact will be achieved via a series of
- Journal papers
- Conference presentations
- Workshops for our industrial partners
- Media publicity where appropriate
The demonstrations will demonstrate to the public the high relevance of this research to society and efficient spending of research funds in terms of the multi-agent vehicle systems. The relevance of this research for networked intelligent autonomous agents with sensing and actuator capabilities for fast response in infrastructure breakdown mitigation in rail transport, nuclear and complex manufacturing systems, will also be demonstrated.
The academic and industrial novelty of the ability of the agents to read system English (sEnglish) technical documents will raise international interest as our agents will read about new physical and mental skills, such as movement controls, movement planning, procedures of payload tasks, situational awareness techniques, response rules and behaviour constraint handling.
Our methods of building up layered abstractions of agent operations, for their physical self-control, communications and team working capabilities, will be academically significant and will show the way for future software engineering in the area of autonomous systems.
The above impact will be achieved via a series of
- Journal papers
- Conference presentations
- Workshops for our industrial partners
- Media publicity where appropriate
Publications
Qu H
(2014)
Formulating Robot Pursuit-Evasion Strategies by Model Checking
in IFAC Proceedings Volumes
Qu H
(2014)
On efficient consistency checks by robots
Qu Hongyang
(2016)
SMCL - Stochastic Model Checker for Learning in Games
in arXiv e-prints
Smyrnakis M
(2014)
Coordination of control in robot teams using game-theoretic learning
in IFAC Proceedings Volumes
Smyrnakis M
(2015)
A Verification Framework for Fictitious Play Based Learning Algorithms
Smyrnakis M
(2016)
Learning of cooperative behaviour in robot populations
Description | Our first result has been on distributed control of frame alignment in multi-vehicle systems. Our second result developed a novel theory for robot cooperation using game theoretic principles which we published and also demonstrated on aerial robots. Our third result provided formal verification of efficient robot coordination in pursuit-evasion multi-robot coordination problems. Our fourth major methodological result has been a human command interface and associated distributed optimisation methods implement on a team of flying vehicles to explore an area in search and rescue tasks in collaboration with our Southampton University Partners. At Sheffield the following has been achieved by the end of May 2013 : 1. Distributed frame alignment problem for multi-agent systems with missing or unreliable attitude sensors has been investigated using camera based computer vision. 2. Possibility of cooperative formation keeping using distributed control and by simple interactions has been proven for large number of spacecraft in theory and simulations. By the end of 2015: 3. Co-active type autonomous cooperation of robots has been developed based on the game theoretic principles of fictitious play 4. Logical consistency of the reasoning and provable performance of the same cooperating robots has been made formally verifiable by new methods developed during the project. 5. Multi-robot and multi-target pursuit-evasion strategies have been developed where cooperating pursuers follow an optimal clearing strategy . By the end of 2017: 6. New combined search and monitoring strategies have been developed as further development of methods in the earlier part of the project. 7. Cooperative agent templates have been developed through the sEnglish Publisher / ROS based integrated development environment for the implementation of practical robotic systems for numerous industrial applications. In 2018, shortly after completion of the project, further results were obtained and published on UAV team based simultaneous search and monitoring of evasive targets on the ground. In 2021 inteliigent agent cooperation policies of dev eloped under this project have been applied to nuclear robotic teamwork. |
Exploitation Route | We are currently working on using the methods in the areas of the nuclear industry and UAVs for cooperative missions. Our methodology and open source software system can enable our partners to adopt the new techniques developed. The new methods support 'transparent robotic' development. |
Sectors | Aerospace Defence and Marine Agriculture Food and Drink Construction Creative Economy Energy Environment Manufacturing including Industrial Biotechology |
URL | https://www.sheffield.ac.uk/acse/research/groups/asrg/acl/project1 |
Description | Academic Impact: Theoretical publications in top journals and conferences on fictitious play based cooperation of robotic systems, pursuit-evasion cooperative strategies, agent templates for autonomous cooperation, verification methods of robot reasoning and of success of cooperation in probabilistic sense. Impact on industrial development: 1. In our ongoing collaboration with the National Nuclear Labs and Sellafield Ltd we have implemented and have demonstrated distributed sensing and computing for cooperative autonomous robot arms for nuclear decommissioning in sort and segregate scenarios. 2. SysBrain Ltd is taking forward the result towards commercial applications of cooperative autonomous robots for the listed industrial sectors. 3. Outcomes of this project are being used in robotic demonstrations to the nuclear industry in 202. |
First Year Of Impact | 2021 |
Sector | Aerospace, Defence and Marine,Agriculture, Food and Drink,Construction,Digital/Communication/Information Technologies (including Software),Energy,Environment,Manufacturing, including Industrial Biotechology,Security and Diplomacy,Transport |
Impact Types | Cultural Societal Economic |
Description | Connected Autonomous Vehicles |
Amount | £73,000 (GBP) |
Organisation | Innovate UK |
Sector | Public |
Country | United Kingdom |
Start | 03/2016 |
End | 05/2017 |
Description | EPSRC: Robotics Capital Funding |
Amount | £1,000,000 (GBP) |
Organisation | Engineering and Physical Sciences Research Council (EPSRC) |
Sector | Public |
Country | United Kingdom |
Start | 03/2014 |
End | 04/2017 |
Description | International Exchanges Scheme |
Amount | £6,000 (GBP) |
Funding ID | IE141180 |
Organisation | The Royal Society |
Sector | Charity/Non Profit |
Country | United Kingdom |
Start | 03/2015 |
End | 05/2016 |
Description | Responsive mode |
Amount | £450,000 (GBP) |
Organisation | Engineering and Physical Sciences Research Council (EPSRC) |
Sector | Public |
Country | United Kingdom |
Start | 11/2014 |
End | 05/2018 |
Description | COSTAIN Group : Autonomous Vehicle Users |
Organisation | Costain Group |
Country | United Kingdom |
Sector | Private |
PI Contribution | 3 days total length of meetings between research staff at Sheffield University and the M1 development team at junction 29 of the M1 motorway. |
Collaborator Contribution | Various robotic needs and development opportunities have been discussed with Costains high level (Tim Embley) and middle management (Mohammed Shah) and their colleagues. |
Impact | The discussions helped us identify research directions and future industrial needs |
Start Year | 2015 |
Description | Floow Ltd: driver experience based learning |
Organisation | Floow Ltd |
Country | United Kingdom |
Sector | Private |
PI Contribution | Providing academic lead in discussion with the company and encouraging them to lead an Innovate UK project proposal where ACSE Sheffield would be the main partner. |
Collaborator Contribution | Guidance on what useful academic/ industrial research could be done to advance progress towards safe an effective autonomous vehicles appearing on our roads. |
Impact | Helped formulate Innovate UK proposal, as partners, which was awarded and its project is due to start shortly. |
Start Year | 2015 |
Description | Meggitt : assistive robots at the work place |
Organisation | Meggitt Aircraft Braking Systems |
Country | United Kingdom |
Sector | Private |
PI Contribution | We have outlined further possibilities of using assistive robots at the manufacturing floor. |
Collaborator Contribution | They described industrial needs and inspired out research direction in the Distributed Sensing Control and Decision Making EPSRC project. |
Impact | Demonstration video, made in our labs at ACSE, which demonstrates two autonomous robot arms working together, which can find objects in the environment, grab them and pass them to each other for further work. |
Start Year | 2015 |
Description | Network Rail |
Organisation | Network Rail Ltd |
Country | United Kingdom |
Sector | Private |
PI Contribution | Meetings between research staff at Sheffield University and a team of technologist from Network Rail, lead by Blacktop |
Collaborator Contribution | Informed us of their application needs for autonomous vehicles in the maintenance of tracks and buildings. |
Impact | We have developed octo-copter drone platforms in the laboratories of the department of ACSE, which are potentially suitable in the canning applications of Network Rail. |
Start Year | 2014 |
Company Name | SysBrain |
Description | SysBrain Ltd provides software consultancy services for engineers and scientists at corporations, institutes, and universities with a focus on operating system software for agent-based control systems using natural language programming. They specialize in the development of robot perception, memory, reasoning, and awareness with their proprietary products. |
Year Established | 2002 |
Impact | The company's software, sEnglish Publisher, has been given with free licence to PDRAs and research students on all EPSRC research projects of the grant holder. This lead to application developments for autonomous surface vehicles with Thales UK, autonomous van with Tata Motors and to research carried out on autonomous drones at the university using the company's software. There was also an EU proposal with the participation of university partners: University of Surrey and University Liverpool where SysBrain Ltd was a key partner. |
Website | http://www.sysbrain.com |
Description | "Intelligent Flying Robots" - University of Sheffield Video, 26th June 2014. |
Form Of Engagement Activity | A press release, press conference or response to a media enquiry/interview |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Media (as a channel to the public) |
Results and Impact | This was a video produced to showcase work on quadcopters undertaken as part of this project. To date it has over 12,500 views on youtube. |
Year(s) Of Engagement Activity | 2014 |
URL | https://www.youtube.com/watch?feature=player_embedded&v=u-j1x-QQuwo |