Environmental issues like pollution are major threats to human health. Many systems are developed to reduce pollution. In this paper, an optimal mobile robot design to reduce pollution in Green supply chain management system. Green supply chain management involves as similating environmentally and economically feasible solutions into the supply chain life-cycle. Smartness, advanced technologies, and advanced networks are becoming pillars of a sustainable supply chain management system. At the same time, there is much change happening in the logistics industry. They are moving towards a new logistics model. In the new model, robotic logistics has a vital role. The reasons for this change are the rapid growth of the e-commerce business and the shortage of workers. The advantages of using robotic logistics are reduction in human errors, faster delivery speed, better customer satisfaction, more safety for workers, and high workforce adaptability. A robot with rocker-bogie suspension is a six-wheeled mobile platform that has a distinctive potential to keep all wheels on the ground continuously. It has been designed to traverse rough and uneven terrain by distributing the load over its wheels equally. However, there is a limitation to achieving high-speed mobility against vertical barriers. In this research, an optimal design of product delivery wheeled robots for a sustainable supply chain system is proposed to ensure higher adaptability and maximum stability during climbing staircases. The design parameters of the proposed robot are optimized using Taguchi Method. The aim is to get a smooth trajectory of the robot’s center-of-mass. The proposed approach realizes a robot with much-improved stability which can climb over heights more than the size of the wheel (i.e., 3 times the radius of wheels). The results reveal that the modified rocker-bogie system not only increases the stair-climbing capability but also thwarts instability due to overturning of a wheel of the robot.
Poor air quality and pollution are causes of diseases and death. According to a recent survey, 9 million deaths per year are caused by pollution. Metro cities and industrial areas are major affected areas in the world. Many research studies tried many solutions for reducing pollution [
The remaining sections of this article are as follow Section 2 reviews the related works about robots with a rocker-bogie suspension system. In Section 3, the mechanism behind conventional rocker-bogie suspension and the overturning problem associated with the conventional system are discussed. A new performance metric, Index of Instability, introduced for smooth running and traversability of some renowned wheeled mobile robots, is assessed using this metric. Section 4 presents the system design for the new rocker-twins bogie mechanism based on the kinematic analyses. In Section 5, the optimization procedure using Taguchi method is discussed and a comparison is made with the achieved optimal robot design with others. From the viewpoints of traversability and smooth running, the extensive experiments using the proposed eight-wheeled robot with the rocker-twins bogies mechanism are carried out in Section 6. At last, Section 7 concludes this research work.
Already, a few researchers have attempted to apply mobile robots to pollution mitigation. Sathiya et al. [
Numerous researchers described how the rocker-bogie mechanism utilised in the Mars Exploration Rover (MER) mission was able to accomplish multiple objectives while emphasising the design’s varied implementations and latch systems [
This section gives information regarding the conventional rocker-bogie mechanism and its drawbacks faced by the conventional system, followed by the proposed stair climbing robot with rocker- bogie mechanism that overcomes the drawbacks seen in the conventional model.
Before designing a new suspension system, a comprehensive kinematic study of suspension mechanism on a rough surface is performed to explore the impacts of lifted wheels against resultant variations in height value and pitch angle (HPA), which is diplomatically used to put forward an innovative stair-climbing product delivery robot. The key challenge of the conventional rocker-bogie suspension systems employed in product delivery robots is their low-speed traversability which derails the rhythm to engross the vibrations produced by wheels shown in To step over vertical barriers, the product delivery robot must be decelerated considerably to provide adequate thrust to lift the mass of the mobile platform. Subsequently, this decreases the average speed of the robot which cannot be accepted in the supply chain management system. If the product delivery robot is moving with high speed and meets a barrier (height more than the diameter of the wheel), there will be a large vibration transferred over the frame which could impair the system or topple down the whole robot.
Even though obstacle dimensions can be different, the most challenging geometry which can be traversed by a product delivery robot is a step or staircase type quadrilateral obstacle shown in
Consequently, from the perspective of smooth-running, it is essential to study the variations of HPA quantitatively. Nevertheless, no appropriate measure for these constraints has been derived since the HPA variations are definitely impacted by many parameters including the type of links implemented, the number of lifted wheels, and the lifting height. Furthermore, as all interactions between lifted wheels and linkages of the robot cannot be taken into account in uneven environments, here lift-up of single wheel is considered to evaluate the traversability of the robot based on variations in HPA. The variables Δ
havg = variation in average height, hstd = standard deviation of variation in height, hmax = maximum height, θavg= average pitch angle variation, θstd = standard deviation of pitch angle variation and θmax = maximum pitch angle
The parameters x, y, and z are used to reflect the impact of each term properly because the robot mobility can be influenced by the numerical value of standard deviation and the maximum variation albeit the average HPA variation is minimum. Firstly, the fundamental assumptions are considered to assess the HPA variations more quantitatively. It is considered that the height of the lift up is the same as the wheel radius. As the radius is the greatest height of any obstacle, a product delivery robot can surmount without using a complex control system to step over the obstacles.
The product delivery robots with classic rocker-bogie mechanism must traverse at a very low speed to guarantee the balanced moving. In the supply chain system, mobile robots may face steps like obstacles to overcome. In this study, a structural adaptation to enlarge the span of classic rocker-bogie suspension mechanism to improve the traversability in retail shops is proposed. An experimental prototype with an additional set of wheels was developed to preserve continuous contact with the ground and to increase frictional force also. The driving mechanism of the new rocker-twin bogies suspension system consists of eight identical wheels, four steering gears, and one differential joint to connect the suspension subsystem with a payload platform, two bogies, two pendulum bars, and two connecting rods as shown in
Furthermore, as the proposed suspension system is actuator-powered, the inclination of the robot can be adapted so that it does not collapse for a large range of slope and enables the robot to traverse over extremely rough surfaces such as steps or staircases. It supplies adequate traction force with the surface even in environments where there is a vertical drop or negative slope of about 1 m by means of a spring-damper system, and it realizes this without negotiating the strength of the body. The wheels, the linear actuator, four-bar linkage, and spring–damper arrangement provide support for the robot to navigate over obstacles.
To assess the performance of a new suspension system to decrease the HPA variations during the stair-climbing process, the kinematic analysis on rugged terrain is performed. The layout of the proposed product delivery robot for kinematic analysis is given in
In the robot front axle, the velocity of wheels’ centers can be defined by the following
Here j = 1, 2, j is the index. It corresponds to left front wheel and right front wheel.
Taguchi method is an efficient parameter-design technique to select an optimum value for all parameters using simulation models or Design of Experiments (DoE) [
As stated earlier, the optimization goal is to change the CM trajectory to a straight line. As the slope of the staircase may differ based on its dimension, it is not rational to find the slope of a straight line. Hence, this optimization objective function is taken as minimization of the area in between CM’s trajectory and a straight line whose slope is just equal to slope of the staircase. The objective function gives equal importance to all sections of CM trajectory. The objective function of this optimization process, includes the straight line, the trajectory of CM, and the equivalent area. These are represented by a red line, a blue line, and a dotted blue line, correspondingly. The straight line’s origin is realized to be a posture of CM, where the front wheel’s moment retains contact with the staircase riser. The rover body’s CM considered in this work includes payloads except for wheels and links. The vehicle’s CM posture varies based on its position during traversing. So, with no generality loss during motion on a staircase, Rover’s CM is fixed to derive a kinematic relationship between wheels and links. In each simulation step, for a small known distance, the rover moves forward on the staircase. Hence, each wheel center’s trajectory is calculated logically from the wheels and links parameters.
In the Taguchi method, S/N ratio is used as the performance metric to calculate the quality of selected link parameters with noise factors. Maximization of the S/N ratio is the prime goal of this optimization. The random noise factors’ effect on the objective function is less when the S/N ratio is high. The goal of this optimization is to reduce the area in between CM’s trajectory and a straight line dictated by each staircase. The S/N ratio is calculated as
where,
Parameters | Values | ||
---|---|---|---|
Stair 1 | Stair 2 | Stair 3 | |
Tread (mm) | 300 | 310 | 240 |
Riser (mm) | 100 | 160 | 200 |
Slope (degree) | 18.4 | 27.3 | 39.8 |
Three different stairs are considered in this study as user conditions. The main aim here is to find the design parameters that give the objective function’s minimum value for the considered types of staircases.
Design parameters are control factors of the system. It is required to find the parameters that affect the system performance. The robot’s successful climbing is the performance to study. Without getting overhung or jammed, the climbing of robots on the staircase is a successful activity. The effective climbing capacity of the robot is studied in this research work. Selecting suitable ranges for control factors is essential to exclude the unfeasible solution space. Different control factors with their choices are defined below:
To begin the Taguchi optimization method, an initial analysis is conducted to estimate the region between three straight lines and the CM trajectory dictated by the suspension system’s predetermined design parameters. Lower and upper limitations apply to the wheel radius. Levels of control parameters and their primary values are given in
Control parameter | Level1(mm) | Level 2(mm) | Level 3 (mm) |
---|---|---|---|
R | 50 | 60 | 70 |
LL 1 | 70 | 80 | 90 |
LL 2 | 130 | 140 | 150 |
LL 3 | 210 | 220 | 230 |
LL 4 | 180 | 190 | 200 |
The Taguchi technique is used to optimise the link dimensions of the rocker-twin bogies system. The process begins with a sensitivity analysis of the S/N ratio. This is to ascertain all key control factors that contribute significantly to the derivation of the objective function. The optimization procedure is then repeated based on these essential control parameters in order to find the optimal values by appropriately espo using a new reduced OA. In
S. No | Control factors | Objective functions | S/N ratio | ||||||
---|---|---|---|---|---|---|---|---|---|
R | LL1 | LL2 | LL3 | LL4 | Y1 | Y2 | Y3 | ||
1 | 50 | 70 | 130 | 210 | 180 | 7210 | 14100 | 16124 | −82.31 |
2 | 50 | 70 | 130 | 210 | 190 | 6867 | 13989 | 14161 | −81.70 |
3 | 50 | 70 | 130 | 210 | 200 | 7519 | 14818 | 13516 | −81.84 |
4 | 50 | 80 | 140 | 220 | 180 | 6700 | 12613 | 13922 | −81.23 |
5 | 50 | 80 | 140 | 220 | 190 | 6857 | 13420 | 14209 | −81.55 |
6 | 50 | 80 | 140 | 220 | 200 | 7253 | 13893 | 14768 | −81.89 |
7 | 50 | 90 | 150 | 230 | 180 | 7645 | 14971 | 17358 | −82.89 |
8 | 50 | 90 | 150 | 230 | 190 | 7126 | 13228 | 14276 | −81.56 |
9 | 50 | 90 | 150 | 230 | 200 | 6658 | 13181 | 14375 | −81.51 |
10 | 60 | 70 | 140 | 230 | 180 | 5918 | 10596 | 11172 | −79.58 |
It is critical to remember that the initial values of the control factors do not reveal kinematic restrictions. Because it is difficult to identify an ideal control factor composition that satisfies kinematic restrictions. Two constraints are violated by this combination of control factors (
R (mm) | LL1(mm) | LL2 (mm) | LL3 (mm) | LL4 (mm) |
---|---|---|---|---|
60 | 70 | 140 | 230 | 180 |
In the second simulation, to satisfy kinematic constraints combined with
To find the factors with the maximum effect on average S/N ratios, the response table is derived as shown in
Level | R | LL1 | LL2 | LL3 | LL4 |
---|---|---|---|---|---|
1 | −81.83 | −81.09 | −81.80 | −81.80 | −81.22 |
2 | −81.35 | −81.54 | −81.37 | −81.39 | −81.72 |
3 | −81.39 | −81.94 | −81.40 | −81.39 | −81.64 |
Delta | 0.48 | 0.85 | 0.43 | 0.41 | 0.50 |
Rank | 3 | 1 | 4 | 5 | 2 |
Control factor | Level 1 (mm) | Level 2 (mm) | Level 3 (mm) |
---|---|---|---|
R | 35 | 40 | 45 |
Link 1 | 90 | 95 | 100 |
Link 4 | 185 | 190 | 195 |
In the second simulation, optimum levels of
S. No | Control factors | Objective functions | S/N ratio | ||||
---|---|---|---|---|---|---|---|
R | LL1 | LL4 | Y1 | Y2 | Y3 | ||
1 | 35 | 90 | 185 | 6287 | 12104 | 12240 | −80.49 |
2 | 35 | 95 | 190 | 6394 | 12005 | 11974 | −80.39 |
3 | 35 | 100 | 195 | 6012 | 10018 | 11104 | −79.38 |
4 | 40 | 90 | 185 | 6543 | 13047 | 12875 | −81.01 |
5 | 40 | 95 | 190 | 6710 | 12071 | 12357 | −80.59 |
6 | 40 | 100 | 195 | 6111 | 11745 | 12476 | −80.43 |
7 | 45 | 90 | 185 | 6624 | 13089 | 12977 | −81.07 |
8 | 45 | 95 | 190 | 6549 | 12147 | 13042 | −80.80 |
9 | 45 | 100 | 195 | 6299 | 12097 | 12674 | −80.63 |
A critical point to remember is that the control factors will be optimised in successive simulations. OAL9 is still used since it is the smallest of the existing OAs for denoting two control elements at three levels. The control factor optimum values for the rocker-twin bogies system are reported in
R (mm) | LL1 (mm) | LL2 (mm) | LL3 (mm) | LL4 (mm) |
---|---|---|---|---|
35 | 100 | 140 | 230 | 195 |
The key motivation of the proposed suspension system is to enable the product delivery robot to give traction for wheels to traverse in high stepped environment. Such a system enables the product delivery robot to travel properly in rugged environments which at present is a challenging endeavor for conventional robots in the supply chain system. In present designs, there is a restriction regarding the height of obstacles which robots can step over. It is very challenging for the conventional wheeled robots to climb up huge vertical obstacles. Using suspension systems with soft shocks, a robot could traverse staircases to an extent but regrettably, it cannot step over very high steps [
Length of links | Wheel Radius | |||
---|---|---|---|---|
LL1 (mm) | LL2 mm) | LL3 (mm) | LL 4(mm) | R (mm) |
100 | 140 | 230 | 195 | 35 |
A wall detecting sensor is used to accurately sense a contact condition during the climbing of stairs. The robot is powered by six high-power 12 V brushed DC motors with metal gears. Motor speed and torque are 180 RPM and 27 kg-cm respectively. The choice of rubber thread attached to the wheel makes it robust and lightweight. It produces excellent friction and traction. These plastic wheels provide a low-cost solution that is light enough to be practical yet still strong enough to operate in rugged terrain. In the prototype, a suitable wheel drive module to fix the wheel hubs inside was considered. It was attached to the steering module using a cantilevered link that aligns the steering axis via the wheel center. Optical encoders were utilized to know the velocity of wheel drive. Absolute magnetic encoders were utilized to know the wheel steer angles. A bogie joint and a rocker joint were attached to the rover chassis’ on both side. Also four steer-drive wheelsets using the links were connected to the rover chassis. With help of a differential mechanism, right-side and left-side suspension systems were joined. Items related to power, computing, communication, and motor controllers were placed in an electronics drive box. One ATMEL Mega 2560 microcontroller was used to do all onboard algorithms related to safety, wireless communication, power management, and driving. Additional software was also used by the microcontroller to enable data logging and debugging facility.
The stair-climbing ability of the proposed system was evaluated with respect to more number of staircases with different widths and heights. The dimensions of stairs and time of traversal are logged for these tests and analysis was carried out to assess the efficiency of the product delivery rover. The objective of the proposed new suspension mechanism is to climb a staircase-like the obstacle of at least three times the wheel radius. The experiments conducted on different staircase dimensions on different surfaces namely wood; ceramic bricks, concrete, and carpet are given in
The speed and IoI of the proposed system are measured while traveling on the rugged terrain. As the value of IoI depends on variation in HPA of the body, a traversal with a low IoI value can be anticipated to demonstrate low HPA variations while traveling in the uneven environment, which ensures smooth movement of the CM shown in
S. No | Type of staircase | Slope(in degree) | Tread (mm) | Riser (mm) | Average Speed (m/s) | IoI |
---|---|---|---|---|---|---|
1 | Concrete | 27.3 | 310 | 160 | 0.2533 | 0.242 |
2 | Brick | 39.8 | 240 | 200 | 0.1438 | 0.224 |
3 | Brick | 18.4 | 300 | 100 | 0.1528 | 0.198 |
4 | Wood | 27.3 | 310 | 160 | 0.2245 | 0.219 |
5 | Wood | 39.8 | 240 | 200 | 0.1367 | 0.268 |
Based on an eminent rocker-bogie mechanism design, this work proposes an optimal design of product delivery wheeled robots in the supply chain system in order to guarantee higher stability and outstanding adaptability during climbing staircases. In the proposed system, the front wheel of the conventional rocker-bogie system is replaced with one more bogie arrangement. The proposed novel eight-wheeled robot with rocker-twin bogies suspension can achieve increased stability and climb over heights more than the chassis height (i.e., 3 times as far as the radius of wheels). The design parameters of the proposed product delivery robot are optimized to get a smooth trajectory for the vehicle’s center-of-mass using the Taguchi method. A new performance measure known as the Index of Instability (IoI) is introduced to assess the smooth running of the robot against a step or staircase. IoI is used to predict unwanted swinging while the robot stepping over the obstacles. This work exhaustively analyzed the traversability of different robots based on IoI metric. Finally, the kinematic analysis is done to demonstrate the traversability of the proposed system over vertical obstacles as related to renowned mobile robots in the literature. The results reveal that the modified rocker-bogie system not only increases stair-climbing capability but also thwarts instability due to overturning of a wheel of the robot. Hence a bright future for staircase robots is available in supply chain industries. Especially for delivering goods to customers can be effectively done by these robots. Visibility of end-to-end delivering goods, automation, and control of order taking and delivery are good using these robots. The advantages of using supply-chain robots are reduced errors, timely delivery, getting timely information, increased trust, efficiency in all working environments, etc. Robots are very reliable for automating both internal and external activities of the supply chain. Suppliers, logistics industries, and customers can rely on the output of supply-chain robots. The performance of the robot during stair climbing is experimentally tested based on the material of the staircase and the degree of slope. Concrete, brick, and wood are the materials considered for the analysis. Different degrees of slope considered here are 27.3, 39.8, and 18.4. For the concrete material with the degree of slope 27.3, the average speed and IoI obtained are 0.2533 and 0.242, respectively. For the wood material with the degree of slope 27.3, the average speed and IoI obtained are 0.2245 and 0.219, respectively. From the above discussed two scenarios, it is seen that when in the degree of slope 27.3 the best average speed is obtained. For the wood material with the degree of slope 39.8, the average speed and IoI obtained is 0.1367 and 0.268, respectively. From the above discussed scenarios, it is seen that the proposed model gives it best with the concrete material and 27.3 degrees of slope.