Low Cost Autonomous Trolley for Labor Assistance in CPEC

Project Title

Low Cost Autonomous Trolley for Labor Assistance in CPEC

Project Area of Specialization RoboticsProject Summary

Summary:

A wide range of human-robot collaborative applications require an autonomous robot to follow its human companion in a diverse environment. China-Pakistan Economic Corridor (CPEC), National Highways Authority (NHA), and CPEC Special Economic Zones (SEZs) require professional labor to accomplish their precise targets and goals. The SEZs, industries, and manufacturing firms are hiring labor for manual handling tasks that may cause a remarkable proportion of labor injuries including work-related musculoskeletal (MSDs) like shoulder dislocation, legs, arms, and joints break, knee injuries, and muscle strains. Morover, recruiting manual labor is costly, time-consuming. It is restricted to normal weather condition.

 To overcome these aboved-mentioned issues, we aim to propose an autonomous labor assistance trolley that will work more robustly and efficiently as compared to human labor. The autonomous trolley will be more vigorous and consistent than humans. Unlike the manual labor massive injuries, the autonomous trolley will be repaired and get back to work without grueling. The autonomous trolley will neither get distracted by the environment nor need to take breaks. It would be effective because the labor assistance trolley will never feel stressed out and stop working, or gossip like humans.

It will work constantly for the whole day, which will speed up the production. There will be no stipulation of allowances like health insurance, wages, food, etc. for the autonomous trolley. CPEC SEZs would be able to maintain their productivity by utilizing these autonomous trollies as they work without any distraction.

Robots are not taking employee's jobs away, they are increasing job opportunities like monitoring, robot mending, supervision. We will build a method that will fabricate the self-autonomous trolley moveable using a two-layer sensory and vision module (monocular vision). If a module collapse, then the other one will replace it to ensure the person-following smooth and steady. From these sensory inputs, the autonomous trolley will extract its drivable path and will also detect the obstacle auspiciously. Then, the motion planner will give commands to robot actuators to follow their assigned person.

Project Objectives

Objectives:

Labor assistance autonomous trolley will make a remarkable change in CPEC SEZs, manufacturing, and industrial fields. It will be able to do its job during the whole day with full potential without any break which will ensure the astonishing change in productivity. We are aiming:

• To assist labor in manufacturing fields.
• To save time, money, and resources.
• To increase the productivity and efficiency of industries, manufacturing fields, and CPEC SEZs.
• It will be the first low-cost labor assistance product of Pakistan.
• It will be “made in Pakistan” product which help to enhance the economic growth of Pakistan.

 An autonomous trolley will be able to achieve the goal on a given platform (CPEC SEZs) to make work more productive more advance. It will require less time to accomplish its assigned task than labor. To increase the production of SEZs and their efficiency, the autonomous labor assistance trolley will play an essential role to achieve the target. Our autonomous trolley will be a vision-based integrated robotic system for real-time application in manufacturing fields that will be useful to increase the working capability of numerous fields. Robotic industrialization, automation, and robotic technologies are a direct source to produce quality work products in industries. The aimed-based autonomous robot with diversity has the potential to yield significant economic, operational, environmental growth. Such robots will be able to exert a real impact on the productivity of special economic zones and inevitably play a growing role in the future. Moreover, Physical injuries such as shoulder and back injuries, etc. among industrial workers may affect the performance of workers, in such conditions autonomous inventions will help to avoid these tragedies.

To compete in the modern world in artificial intelligence technology our autonomous trolley will increase productivity and reliability as its modules will be designed according to modern and efficient technologies that help us to achieve our working aims and goals.

Project Implementation Method

Implementation:

 The detailed implementing method for the autonomous labor-assistance trolley is categorized as

Data acquisition:

Implementing methods play a vital role in the exultant production of any specific module. For this purpose, we are collecting indoor and outdoor person dataset from different cities of Pakistan. This process is preferential because the labor in our culture has a different appearance as compared to western culture. It is expected to collect a dataset of more than 15000 images. It will help to train our model in a deep learning algorithm for the autonomous trolley.

Core Modules:

The autonomous trolley will be consisting of two core models.

• Hardware
• Software

Hardware:

The hardware component consist of a monocular vision HD camera, ultrasonic sensor, microprocessor, and power-up batteries. The monocular HD camera will be used to capture the live video stream. Besides, a sonar sensor will also be used to detect the obstacle in severe illusion and occlusion where the camera unable to detect the obstacle meticulously. The microprocessor will be used to process the sensory information(shown in figure 1). While batteries will be used to power-up the system.

Software:

The software module will be consisting of a perception and motion planning module.

PERCEPTION Module:

The perception module will be trained on a deep learning algorithm to perceive the environment in real-time. Once the model will be trained the autonomous trolley will be able to detect and classify the intended person as well as to detect the obstacles in video stream captured through the camera. The perceptual information will be conveyed to the motion planning module (as shown in figure 1).

Motion Planner Module:

The motion planner will be responsible for the extraction of the drivable area. It will  give commands to the actuator(as shown in figure 1 ) which will control the acceleration, braking, and steering of the autonomous trolley . Simultaneously, the motion planner will also be responsible to perpetuate a particular distance from the targeted person, object or vehicle.

           Figure1. Flow Diagram of Autonomous Trolley

Benefits of the Project

Benefits:

Benefits of Labor assistant trolley seem to be most visible in the following terms:

Productivity:

These labor assistance robot hand over labor tasks by providing industries precise and high-quality work than manual labor. Where manual labor is tired because of their physical limitations these robots persistently work without any lunch break, a day off, or sick leave. They only need breaks when they need to repair and upgrading. They also carry out construction material more repeatable than manual labor. Many industries need these robot's assistance for their great innovations.

Safety:

The manifest primacy of utilizing labor robot is safety. These robots carry material to prevent humans from serious injuries like muscle strains, Back strain, ligament sprains, neck, and spin stress. Heavy material, high temperature of steal danger to hold, sharp object that easily injures labor. By deputing these riskiness tasks to these labor assistant robot industries can look at repair cost rather than medical cost.

Cost-effectiveness:

Another benefit of the Labor assistance robot is that it saves time by being able to perform a massive number of tasks. These robots save the money of industrial companies in the long run with quick RIOs by recruiting fewer workers, zero injuries cost. These robots work on a repetitive cycle as long as drained or need repairing because of this companies can recover their investment in a short amount of time Because these robots can work 24/7.

We can also use these robots as personal assistants to haul luggage from airports, hotels, warehouses, grocery stores, or even in the hospital for carrying medical equipment. Humans can walk around freely in parks, gyms, etc. because these robots haul their stuff for them.

Technical Details of Final Deliverable

Technical Details of Final Deliverable:

The autonomous labor assistance trolley will rely on a monocular camera, actuators, ultrasonic sensors, complex algorithm, machine learning system, and a powerful processor to execute its autopilot. The sophisticated software of this autonomous trolley will process all sensory input, plan a path, and will send the instruction to autopilot for autonomous navigation. This software will be supported by the person classification and following model, predictive modeling, obstacle detection, and avoidance algorithm.

Our model will present a vision-based human-following strategy. The autonomous trolley will be equipped with a monocular vision-based HD camera to capture the live stream. The visual data will be received as input through live streaming that will proceed in a deep learning module to detect and classify the object in the frames. The module will also help to detect the drivable area for the autonomous trolley in real-time. After the identification and classification of the intended person, the motion planner module will send commands to actuators that will be responsible to control acceleration, braking, and steering. It will also assist the autonomous trolley to maintain explicit distance from the focused person. While chasing, the trolley would be able to detect the obstacle and avoid getting clash with it. In the meantime, it will find the left or right drivable area and the appropriate direction will be chosen to continue the specific person-succeeding in real-time.

Sometimes camera may deplete to detect the obstacle in severe illusion and occlusion. To overcome this glitch, we will use the ultrasonic sensor to get better ramifications. With the use of an ultrasonic sensor, the trolley will ensure the smooth run-down of its determined target even in severe illusion and occlusion.

Final Deliverable of the Project Hardware SystemCore Industry TransportationOther IndustriesCore Technology RoboticsOther TechnologiesSustainable Development Goals Decent Work and Economic Growth, Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	80000
HD Color Camera	Equipment	1	25000	25000
Ultrasonic Sensor	Equipment	2	500	1000
Microcontroller	Equipment	1	30000	30000
Chassis and Foundry	Equipment	1	5000	5000
Dry Cell Battery	Equipment	1	9000	9000
Contingencies	Miscellaneous	1	3000	3000
Visits	Miscellaneous	1	3000	3000
Services	Miscellaneous	1	3000	3000
Stationary	Miscellaneous	1	1000	1000