Multi-Parameter Predictive Model of Mobile Robot’s Battery Discharge for Intelligent Mission Planning in Multi-Robot Systems
Abstract
:1. Introduction
2. Materials and Methods
2.1. Environmental Key Mission Parameters
2.2. Key Mission Parameters
2.3. Resistance to Motion
2.4. Power Source
2.5. Available Status Parameters
2.6. Measuring System and Communication Protocols
Algorithm 1: Data acquisition and analysis from each mission |
WHILE true: mission ← getMissionID() currentMission ← mission initialStatus ← getRobotStatus() WHILE mission == currentMission: status ← getRobotStatus() avgParams ← countAvgParams(status.linV, status.angV, status.batTemp) currentMission ← getMissionID() ENDWHILE statusDifference ← calcDifference(status, initialStatus) missionParams ← calcMissionParams(avgParams, statusDifference) DataBase ← (missionParams, avgParams) ENDWHILE |
2.7. Data Acquisition
3. Results
- G1.
- Identification of the parameters that have the greatest impact on the characteristics of energy consumption of Lithium NMC batteries used in the MiR100 autonomous mobile robot.
- G2.
- Finding the optimal model for predicting the consumption of the battery charge level used in the MiR100 autonomous mobile robot based on the key parameters.
- G3.
- Comparison of the obtained models for two batteries from the same manufacturer that differ in the production date and the total exploitation time.
- Forward selection;
- Stepwise regression;
- Least absolute shrinkage and selection operator (LASSO);
- Least angle regression (LAR).
- filter 1 (EneryCons > 0.2);
- filter 2 (SoC < 80 and SoC > 20);
- filter 3 (SoC < 80 and SoC > 20 and EnergyCons > 0.2).
4. Discussion
- R1: Having specified the potential parameters influencing the battery consumption in a single mission and using the data collected from multiple battery discharge cycles under various conditions, a forward selection algorithm was used to find the optimal parameters to formulate the model of battery discharge. After the analysis, it is clear that the parameters with the highest impact on the battery discharge are the current SoC and the level of complexity of the mission itself, which can be described by the distance to travel and the number of turns.
- R2: Having analysed different modelling algorithms combined with different data filters, an optimal modelling algorithm—forward selection algorithm—was chosen. Upon further analysis of the parameters of the model, the authors have decided to limit the parameters to the first four parameters of the model since they were the same for both of the tested batteries, and including additional parameters would not significantly increase the AR-S value, which had been chosen as the indicator of the model’s performance. The AR-S value for the first four parameters was equal to 0.95 while including more parameters could increase this value up to 0.96 (see Table 5). From a practical point of view, the authors suggest using the simpler model for the predictive model of battery discharge. The independent variables used in the model are the number of turns (Turns), the travel distance (Distance), the current SoC of the battery (SoC) and the interaction of variables SoC * Distance. The factors of these variables for both of the tested batteries can be found in Table 6.
- R3: The experiments conducted for both of the batteries have yielded similar results in terms of the most prominent independent variables as well as their factors influencing the battery discharge (see Table 6), which leads us to a conclusion that, in the case of the two tested batteries, the level of exploitation and the age of the battery does not affect the model, though only two batteries do not provide enough evidence and further, more extended tests should be conducted for more units of batteries to verify this claim. Since the models for both batteries are nearly identical, either one of them can be used for predicting the battery discharge in any mission designed for the MiR100 autonomous mobile robot.
5. Summary and Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Sun, X.; Hao, Y.; Solvang, W.D. Towards the Smart and Sustainable Transformation of Reverse Logistics 4.0: A Conceptualization and Research Agenda. Environ. Sci. Pollut. Res. 2022, 29, 69275–69293. [Google Scholar] [CrossRef] [PubMed]
- Sung, I.; Nam, H.; Lee, T. Scheduling algorithms for mobile harbor: An extended m-parallel machine problem. Int. J. Ind. Eng. Appl. Pract. 2013, 20, 211–224. [Google Scholar]
- Fang, X.L.; Wang, Y.B.; Dong, W.; Yang, Q.; Sun, S.Y. Optimal Energy Management of Multiple Electricity-Hydrogen Integrated Charging Stations. Energy 2023, 262, 125624. [Google Scholar] [CrossRef]
- Stączek, P.; Pizoń, J.; Danilczuk, W.; Gola, A. A Digital Twin Approach for the Improvement of an Autonomous Mobile Robots (AMR’s) Operating Environment—A Case Study. Sensors 2021, 21, 7830. [Google Scholar] [CrossRef] [PubMed]
- Anczarski, J.; Bochen, A.; Głąb, M.; Jachowicz, M.; Caban, J.; Cechowicz, R. A Method of Verifying the Robot’s Trajectory for Goals with a Shared Workspace. Appl. Comput. Sci. 2022, 18, 37–44. [Google Scholar] [CrossRef]
- Sen Sarma, D.; Warendorf, T.; Espin-Sarzosa, D.; Valencia-Arroyave, F.; Rehtanz, C.; Myrzik, J.; Palma-Behnke, R. Multi-objective Energy Management for Modern Distribution Power Systems Considering Industrial Flexibility Mechanisms. Sustain. Energy Grids Netw. 2022, 32, 100825. [Google Scholar] [CrossRef]
- Rahman, H.F.; Janardhanan, M.N.; Nielsen, P. An integrated approach for line balancing and AGV scheduling towards smart assembly systems. Assem. Autom 2020, 40, 219–234. [Google Scholar] [CrossRef]
- Bocewicz, G.; Nielsen, I.; Gola, A.; Banaszak, Z. Reference model of milk-run traffic systems prototyping. Int. J. Prod. Res. 2021, 59, 4495–4512. [Google Scholar] [CrossRef]
- Fragapanea, G.; Kosterb, R.; Sgarbossaa, F.; Strandhagena, J.O. Planning and Control of Autonomous Mobile Robots for Intralogistics: Literature Review and Research Agenda. Eur. J. Oper. Res. 2021, 294, 405–426. [Google Scholar] [CrossRef]
- Mei, Y.; Lu, Y.-H.; Hu, Y.C.; Lee, C.S.G. A Case Study of Mobile Robot’s Energy Consumption and Conservation Techniques. Available online: https://ieeexplore.ieee.org/document/1507454 (accessed on 13 October 2022).
- Zhao, X.; Osborne, M.; Lantair, J.; Robu, V.; Flynn, D.; Huang, X.; Fisher, M.; Papacchini, F.; Ferrando, A. Towards Integrating Formal Verification of Autonomous Robots with Battery Prognostics and Health Management. Softw. Eng. Form. Methods 2019, 11724, 105–124. [Google Scholar] [CrossRef]
- Tomy, M.; Lacerda, B.; Hawes, N.; Wyatt, J.L. Battery Charge Scheduling in Long-Life Autonomous Mobile Robots via Multi-Objective Decision Making under Uncertainty. Rob. Auton. Syst. 2020, 133, 103629. [Google Scholar] [CrossRef]
- Krot, K.; Iskierka, G.; Poskart, B.; Gola, A. Predictive Monitoring System for Autonomous Mobile Robots Battery Management Using the Industrial Internet of Things Technology. Materials 2022, 15, 6561. [Google Scholar] [CrossRef] [PubMed]
- Rappaport, M.; Bettstetter, C. Coordinated Recharging of Mobile Robots during Exploration. Available online: https://ieeexplore.ieee.org/document/8206600 (accessed on 13 October 2022).
- Oliveira, M.M.; Galdames, J.P.M.; Vivaldini, K.C.T.; Magalhães, D.V.; Becker, M. NICD Battery Discharging Estimation System for AGVs Working in Intelligent Warehouses Based on EKF. In ABCM Symposium Series in Mechatronics; Alfaro, S.C.A., Motta, J.M.S., De Negri, V.J., Eds.; ABCM—Brazilian Society of Mechanical Science and Engineering: Rio de Janeiro, RJ, Brasil, 2012; Volume 5, pp. 1122–1131. [Google Scholar]
- Partovibakhsh, M.; Liu, G. An Adaptive Unscented Kalman Filtering Approach for Online Estimation of Model Parameters and State-of-Charge of Lithium-Ion Batteries for Autonomous Mobile Robots. Available online: https://ieeexplore.ieee.org/document/6814843 (accessed on 13 October 2022).
- Lacerda, B.; Faruq, F.; Parker, D.; Hawes, N. Probabilistic Planning with Formal Performance Guarantees for Mobile Service Robots. Int. J. Rob. Res. 2019, 38, 1098–1123. [Google Scholar] [CrossRef]
- Gravesteijn, B.Y.; Nieboer, D.; Ercole, A.; Lingsma, H.F.; Nelson, D.; van Calster, B.; Steyerberg, E.W. Machine Learning Algorithms Performed No Better than Regression Models for Prognostication in Traumatic Brain Injury. J. Clin. Epidemiol. 2020, 122, 95–107. [Google Scholar] [CrossRef] [PubMed]
- Aguelo, B.O.; Zamboni, W.; Postiglione, F.; Monmasson, E. Battery State-of-Health Estimation Based on Multiple Charge and Discharge Features. Energy 2023, 263, 125637. [Google Scholar] [CrossRef]
- Chen, J.; Yang, C.; Zhou, C. Multivariate Regression Model for Industrial Process Measurement Based on Double Locally Weighted Partial Least Squares. Available online: https://ieeexplore.ieee.org/document/8863943 (accessed on 13 October 2022).
- Yang, Y.; Campus, Q. Naval Aviation University, Qingdao, China Prediction and Analysis of Aero-Material Consumption Based on Multivariate Linear Regression Model. Available online: https://ieeexplore.ieee.org/document/8386591 (accessed on 13 October 2022).
- Li, D. Predicting Short-Term Traffic Flow in Urban Based on Multivariate Linear Regression Model. J. Intell. Fuzzy Syst. 2020, 39, 1417–1427. [Google Scholar] [CrossRef]
- Saluci, C.B.; Bakdi, A.; Glad, I.K.; Vanem, E.; De Bin, R. Multivariable Fractional Polynomials for Lithium-Ion Batteries Degradation Models under Dynamic Conditions. J. Energy Storage 2022, 52, 104903. [Google Scholar] [CrossRef]
- Sun, S.; Zhang, J.; Bi, J.; Wang, Y. A Machine Learning Method for Predicting Driving Range of Battery Electric Vehicles. J. Adv. Transp. 2019, 2019, 4109148. [Google Scholar] [CrossRef]
- Abdelaty, H.; Mohamed, M. A Prediction Model for Battery Electric Bus Energy Consumption in Transit. Energies 2021, 14, 2824. [Google Scholar] [CrossRef]
- Liu, Z.Y.; Xie, J.; He, H.J.; Wang, K.Q.; Huang, W. Prediction Method for Battery Self-Discharge Voltage Drop Based on Pre-Classifier. Measurement 2022, 204, 112065. [Google Scholar] [CrossRef]
- Liu, Z.Y.; He, H.J.; Xie, J.; Wang, K.Q.; Huang, W. Self-Discharge Prediction Method for Lithium-ion Batteries Based on Improved Support Vector Machine. J. Energy Storage 2022, 55, 105571. [Google Scholar] [CrossRef]
- Conte, C.; Rufino, G.; de Alteriis, G.; Bottino, V.; Accardo, D. A Data-Driven Learning Method for Online Prediction of Drone Battery Disharge. Aerosp. Sci. Technol. 2022, 130, 107921. [Google Scholar] [CrossRef]
- Zhao, D.; Zhou, Z.J.; Zhang, P.; Zhang, Y.J.; Qin, H.B.; Gao, S. End-of-Discharge Prediction for Sattelite Lithium-ion Battery Based on Evidential Reasoning Rule. Open Astron. 2022, 31, 256–267. [Google Scholar] [CrossRef]
- Gockcen, A.; Gocken, A.; Sahin, S. Prediction of Li-Ion Battery Discharge Patterns in IoT Devices Under Random Use Via Machine Learning Algorithms. Comput. J. 2022, 39, 84–90. [Google Scholar] [CrossRef]
- Zhao, H.C.; Wang, X.R.; Bai, Y.; Jin, H.F.; Zhou, J.; Tan, G.Q.; Wu, C. Thermal simulation and prediction of high-energy LiNi0.8Co0.15Al0.05O2//Si-C pouch battery during rapid discharging. J. Energy Storage 2022, 47, 103356. [Google Scholar] [CrossRef]
- Zou, L.; Wen, B.Y.; Wei, Y.Y.; Zhang, Y.; Yang, J.; Zhang, H. Online Prediction for Remaining Useful Life for Li-Ion Batteries Based on Discharge Voltage Data. Energies 2022, 15, 2237. [Google Scholar] [CrossRef]
- Yin, A.J.; Tan, Z.B.; Tan, J. Life Prediction Battery Using a Neural Gaussian Process with Early Discharge Characteristics. Sensors 2021, 21, 1087. [Google Scholar] [CrossRef]
- Mobile Industrial Robots 2.7.1 MIR100 REST API Documentation. Available online: https://www.jugard-kuenstner.de/fileadmin/daten/Downloads/Intralogistik/MiR_Transportsystem/MiR100_MiR200/MiR_Rest-API.pdf (accessed on 13 October 2022).
- Kostopoulos, E.D.; Spyropoulos, G.C.; Kaldellisa, J.K. Real-World Study for the Optimal Charging of Electric Vehicles. Energy Rep. 2020, 6, 418–426. [Google Scholar] [CrossRef]
- Eider, M.; Deggendorf, A.B. Institute of Technology, Deggendorf, Germany Dynamic EV Battery Health Recommendations. Available online: https://www.researchgate.net/publication/325217548_Dynamic_EV_Battery_Health_Recommendations (accessed on 13 October 2022).
- Wang, Z.; Wang, Y.; Li, Z.; Song, Q.; Rong, Y. The Optimal Charging Method Research for Lithium-Ion Batteries Used in Electric Vehicles. Energy Procedia 2016, 104, 74–79. [Google Scholar] [CrossRef]
- Schocha, J.; Gaerttnera, J.; Schullerb, A.; Setzerc, T. Enhancing Electric Vehicle Sustainability through Battery Life Optimal Charging. Trans. Res. Part B Methodol. 2018, 112, 1–18. [Google Scholar] [CrossRef]
Measurable Parameters | Irrelevant, Constant or Unpredictable Parameters |
---|---|
|
|
Payload | Battery 1 | Battery 2 | ||||
---|---|---|---|---|---|---|
30 m | 140 m | 350 m | 30 m | 140 m | 350 m | |
0 kg | 2 | 2 | 2 | 3 | 4 | 2 |
50 kg | 2 | 4 | 3 | 3 | 3 | 3 |
100 kg | 3 | 3 | 3 | 4 | 3 | 4 |
Description of Variable | Label | Type of Variable |
---|---|---|
Measured energy consumption | EnergyCons | dependent/response |
Date of measurement | Date | description/not used |
Battery temperature | BatteryTemp. | independent/predictor |
Battery SoC | SoC | independent/predictor |
Average linear velocity | AvgLinVelocity | independent/predictor |
Average angular velocity | AvgAngVelocity | independent/predictor |
Number of turns in a mission | Turns | independent/predictor |
Payload | Payload | independent/predictor |
Average mission travel distance | Distance | independent/predictor |
Model Selection | Battery 1 | Battery 2 | ||||||
---|---|---|---|---|---|---|---|---|
No Filter | F1 | F2 | F3 | No Filter | F1 | F2 | F3 | |
Forward | 0.9399 | 0.9466 | 0.9568 | 0.9629 | 0.9489 | 0.9540 | 0.9646 | 0.9694 |
Stepwise | 0.9399 | 0.9467 | 0.9568 | 0.9621 | 0.9489 | 0.9541 | 0.9645 | 0.9694 |
LASSO | 0.9355 | 0.9407 | 0.9514 | 0.9523 | 0.9399 | 0.9461 | 0.9540 | 0.9601 |
LAR | 0.9387 | 0.9407 | 0.9527 | 0.9523 | 0.9469 | 0.9487 | 0.9650 | 0.9693 |
Step | Battery 1 | Battery 2 | ||
---|---|---|---|---|
Effect Entered | Adjusted R-Square | Effect Entered | Adjusted R-Square | |
1 | Turns | 0.9139 | Turns | 0.9167 |
2 | Distance | 0.9196 | Distance | 0.9202 |
3 | SoC | 0.9271 | SoC | 0.9297 |
4 | SoC * Distance | 0.9508 | SoC * Distance | 0.9591 |
5 | Distance * Distance | 0.9521 | SoC * SoC | 0.9599 |
6 | SoC * SoC | 0.9530 | BatteryTemp | 0.9655 |
7 | AvgAngVelocity | 0.9565 | Distance * Distance | 0.9662 |
8 | SoC * SoC * SoC | 0.9573 | SoC * Turns | 0.9667 |
9 | AvgAngVel * AvgAngVel | 0.9613 | BatteryTemp * SoC | 0.9670 |
10 | SoC * Turns | 0.9620 | SoC * SoC * SoC | 0.9679 |
11 | Payload | 0.9624 | BatteryTe * BatteryTem | 0.9681 |
12 | AvgAngVel * Distance | 0.9626 | BatteryTemp * Distance | 0.9689 |
13 | AvgAng * AvgAng * AvgAng | 0.9628 | Payload | 0.9691 |
14 | SoC * Payload | 0.9628 | Payload * Payload | 0.9692 |
15 | Turns * Payload | 0.9629 | AvgLinVelocity | 0.9693 |
16 | AvgLinVel * AvgLinVel | 0.9693 | ||
17 | Turns * Distance | 0.9694 |
Battery 1 | Battery 2 | ||
---|---|---|---|
Effect | Effect Factor | Effect | Effect Factor |
Turns | −0.027318 | Turns | −0.031285 |
Distance | 0.002327 | Distance | 0.002999 |
SoC | −0.141574 | SoC | −0.136106 |
SoC * Distance | 0.000000246 | SoC * Distance | −0.000037981 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Poskart, B.; Iskierka, G.; Krot, K.; Burduk, R.; Gwizdal, P.; Gola, A. Multi-Parameter Predictive Model of Mobile Robot’s Battery Discharge for Intelligent Mission Planning in Multi-Robot Systems. Sensors 2022, 22, 9861. https://doi.org/10.3390/s22249861
Poskart B, Iskierka G, Krot K, Burduk R, Gwizdal P, Gola A. Multi-Parameter Predictive Model of Mobile Robot’s Battery Discharge for Intelligent Mission Planning in Multi-Robot Systems. Sensors. 2022; 22(24):9861. https://doi.org/10.3390/s22249861
Chicago/Turabian StylePoskart, Bartosz, Grzegorz Iskierka, Kamil Krot, Robert Burduk, Paweł Gwizdal, and Arkadiusz Gola. 2022. "Multi-Parameter Predictive Model of Mobile Robot’s Battery Discharge for Intelligent Mission Planning in Multi-Robot Systems" Sensors 22, no. 24: 9861. https://doi.org/10.3390/s22249861
APA StylePoskart, B., Iskierka, G., Krot, K., Burduk, R., Gwizdal, P., & Gola, A. (2022). Multi-Parameter Predictive Model of Mobile Robot’s Battery Discharge for Intelligent Mission Planning in Multi-Robot Systems. Sensors, 22(24), 9861. https://doi.org/10.3390/s22249861