Multi Agent Plan Excecution and Monitoring for Service Robotics
Multi-agent collaboration in static environments has been a hot research area that has witnessed incredible advances for the last decade and with the very notion, we want to develop a High-level planner that allocates tasks to multiple agents working on a Low-level in house environment with both sta
| Project Title |
Multi Agent Plan Excecution and Monitoring for Service Robotics
| Project Area of Specialization |
Robotics | | Project Summary |
Multi-agent collaboration in static environments has been a hot research area that has witnessed incredible advances for the last decade and with the very notion, we want to develop a High-level planner that allocates tasks to multiple agents working on a Low-level in house environment with both static and movable objects to keep the house Clean and Tidy. Issues like commonsense knowledge (e.g. robots should know to put a dirty dish lying on the table in the dishwasher and a book on the bookshelf ), agent collaboration (e.g. a robot can call another one to help in case a heavier object is to be lifted) and task recovery planning in case of failure (e.g. the object to be moved have been misplaced by an entity outside the scope of the planner such that a human) will be addressed through utilization of Action languages like C+, STRIPS and ASP(answer set programming) and low-level plan will be implemented through Robot Operating System, ROS. Our solution can be extended to any dynamic one. | | Project Objectives |
- Research analysis on developmemt of High level plan for service robotics and its integration at low level to develop robot house keeping assitants in real time for a sample of 20 scenarios.
- Multi Agent task planning.
- Multi Robot collaboration in static environment.
- Multi Agent autonomous navigation.
- Robot Motion and task path planning.
| | Project Implementation Method |
- Building of high level domain and problem scenarios in C+ action language through CCALC.
- Planing a solution with in ZCHAFF solver.
- Integration of the high level plan with ROS (noetic) through Python3.
- Utiliztion of SLAM algorithm for autonomous nagivigation.
- Following similar strategy developing solutions for 20 scenarios and their implementions to build a research model.
| | Benefits of the Project |
- In multi agents task planing we are able to fulfil our objective through a network of interconnected agents with optimal efficiency and in minimum time by providing them computational resources.
- As multiple agents working is the inter-operation of several legacy systems so they will give the better outcome as whole.
- As our multiple agents are autonomous interacting agents, therefore our system has quite natural approach to get a goal and user friendly.
- By dividing the expertise spatially and temporally these agents will help the mankind at the point where human beings will face difficulty.
- In our project the multiple agents help us to achieve a single goal that is cleaning with no human effort.
- A person will be able to give more time and focus to his office work as he or she will not have any household chores to manage.
| | Technical Details of Final Deliverable |
- Basically we are doing the research work which enables us to gather the maximum information and details regarding how to perform a task through multiple agents in a particular world(environment) through high level planning.
- After selecting and creating our sample domain ,that is a household environment and different planning scenarios we will be developing the task plan through high level planner and its integration at low level on ROS.
- Our project will be a success if we are able to execute the plans properly to get the optimal result from the multiple agents which are being monitored.
- The research work will give a lay man enough information from the basics to the higher level, that makes him capable of using the multiple agents(robots) according to his own will with the optimal efficiency in his environment(of any type) and provide the basis for multi agent planning alorithms.
| | Final Deliverable of the Project |
Software System | | Core Industry |
Others | | Other Industries |
IT | | Core Technology |
Robotics | | Other Technologies |
Artificial Intelligence(AI) | | Sustainable Development Goals |
Good Health and Well-Being for People, Industry, Innovation and Infrastructure | Required Resources
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
| Intel RealSense Depth Camera D415 | Equipment | 1 | 35000 | 35000 |
| Nvidia geforce Gtx 1650 Super | Equipment | 1 | 35000 | 35000 |
| Documentation | Miscellaneous | 1 | 10000 | 10000 |
| | | Total in (Rs) | 80000 |