Internet of Things based Speed Control for an Industrial Electric Vehicle using ARM Core

Increasing greenhouse gases impose severe concern over the environment resulting in rising dangerous calamities of climate change in the form of floods, etc. Major disadvantages like intermittency of electric vehicles need to be charged after traveling fixed distance. The paper develops an algorithm for a selected industrial electric vehicle to be controlled at different speeds that envisages working on real time Internet of Things (IoT) based Global Positioning System (GPS) signals. It engages the ARM core based STM micro-controller in conjunction with mesh networked Bluetooth Low Energy (BLE) to govern the operations besides enabling it to be dynamically monitored. The system design considers the vehicle parameters that include the speed of vehicle and the engine, State of Charge (SoC) and State of Health (SoH) of battery together with real time GPS based navigation system using IoT bundled GPS based maps interface. The methodology involves a closed loop monitoring with specified sequence of steps that augur the system to operate at defined speed over designated work shifts and schedules. The procedure introduces an embedded C environment with a process of unit-testing based simulation to capture the merits of schema in terms of an improved vehicle performance under varying parametric conditions.


INTRODUCTION
The industrial vehicles continue to gather momentum owing to the extensive automation prevalent in both on and off-road goods category.However, in view of the rapidly depleting conventional resources, there is a strong need to explore the use of electric powered vehicles and comply with the clean energy initiatives.In addition there is the urge for the vehicles to be smarter, safe, and energy efficient for sustaining a longer operational period with lower maintenance and running cost, being amply supported by increased vehicle efficiency with smaller downtime.
The transportation sector appears to experience a greater push to become electric and support goods vehicle due to increased traffic movement.It mandates the vehicles to operate at optimum capacities for sustaining maximum efficiency alongside operating the vehicles within its speed limit without affecting the mileage by consuming less fuel for the same distance covered.
The industrial electric vehicles necessitate the vehicles to operate at a recommended speed to ensure the sustenance of the life of the battery for longer tenure.The intelligent autonomous mode can be aided by one or more systems that monitor the vehicle continuously and control its speed within the prescribed limits.However, with the manual mode the speed control depends on the driver which may be strenuous, considering the quantum of workload.
The vehicle segment encompasses a significant space in view of the exhaustive automation in almost every class of both on and off-road passenger and goods traffic.The introduction of the embedded segment brings in an impetus enabling wireless vehicle-to-vehicle communications system.
The driverless vehicles have been a part of the Intelligent Transportation Systems (ITS) with the release of several intelligent vehicles to the roads which in turn necessitated the coordination among the vehicles and the associated infrastructure (Turcian and Dolga 2020).
The architecture for using three diff erent categories of Vehicle to Everything (V2E) communications schemes in off -road environments over multiple autonomous vehicles has been proposed.Numerous experiments have been performed for each communication scheme and the results prove the stability of the schemes in addition to establishing their high performance over 4G connection in terms of effi ciency, in comparison to Wi-Fi connection (Al-Sultan et al., 2014; Schweppe et al., 2011).
The autonomous industrial off -road vehicles that include baggage tractors, drone-based surveillance systems, and vehicles in mines or construction sites have been operated and proved the ease of design with a higher degree of control The emphasis relates to evolving an algorithm for controlling the speed of an industrial electric vehicle mostly operated manually over longer duration to ensure better payload movement effi ciency.It attempts to use mesh BLE network to operate the vehicle from an IoT platform using real time GPS tracking.The design orients to encompass the vehicle parameters that include the speed of the vehicle and the engine, SoC and SoH of battery for arriving at the predefi ned speed over varying parametric uncertainties.It incites to follow a unit testing approach on embedded C environment for demonstrating the claim to be used in practice.

PROPOSED WORK
The off -road electric vehicle faces its own challenge depending on the usage and its area of application.The industrial electric vehicles often operate over the recommended speed to move more goods within a stipulated shift timing, reducing the life of the vehicle and the battery.
Operating the vehicle at the optimum speed can provide the much-needed life for the vehicle and safety to the people who move around the vehicle with in a predefi ned closed industrial premises.
Figure 1 explains the high-level design including the battery-operated electric power to the vehicle and its dependent circuits articulating the basic approach to control the speed across vehicles parametric considerations.Designing a controllable track able Intelligent vehicle to maintain speed can help fl eet owners to extract maximum effi ciency.
The industrial vehicle under study with technical specifi cations in Table 1 may reach a top speed of 25 km/hr and can be expected to be operated on a flat surface within an industrial complex operated using a sealed Lead Acid Battery.
The parameters to be monitored include the Battery SoC, Battery SoH, Driver Fatigue (Eye) Detection, Real Time Navigation points using GPS, Vehicle Speed, Motor Temperature, Motor Current Consumption, Tire Pressure and Truck Payload.The efforts to control the speed require monitoring the parameters closely at a constant periodic interval across the complete vehicle, including idle time at millisecond accuracy maintained by a RTOS.Table 2 records a sample of the parametric values for the chosen vehicle.

METHODOLOGY
The primary exercise endeavours to form a BLE Mesh network around the area within which the vehicle envisages to move or designated to traverse with safe warnings including audio and visual with flashing lights.The network interconnects the mesh within the traffic space and further to the centralized on-premises data collection centre for monitoring and control with the vehicle with a BLE device in it.It operates on the real-time data from the different sensors interfaced to the STM32F746NG micro-controller.
The main objective revolves around close vehicle monitoring on constant periodic intervals to trigger the control logic in the controller to decide the specified speed depending on the parameters under consideration, in addition to the pay load in the vehicle.
It brings in a closed loop control system approach with the monitoring parameters forming the input, the micro-controller edge device smartness acting as the process block and the speed control acting as the output, with the vehicle speed also acting as a feedback signal together with appropriate signal conditioning circuits.The firmware code in the embedded-c framework with RTOS allows the formulation in the unit testing at bench.STM32F746NG communicates with all sensors in the vehicle and collects the data.The IoT monitoring and control interface layer incorporates the BLE protocol to communicate between the other BLE nodes installed across the premises.The stationary BLE nodes act as repeaters and extenders, carry the vehicle data from the vehicle to the centralized data collection master and there from returns the control information back to the vehicle.The smart edge device powered by STM-32F746NG inherits inbuilt intelligence to control the speed of the vehicle based on the dynamic behaviour of the vehicle and its parameters.
The process enables the interface of the Bluetooth Module HC-05 with the STM on SPI (Serial Peripheral Interface) and includes BLE 4.0 stack in it with additional application layer implementation.The stationary BLE nodes installed in the industry can be an off -the-shelf certifi ed interoperable customizable repeater to facilitate communication with the centralized on-premises data collection unit.
It identifi es the centralized on-premises unit by a static addressing mechanism to transmit the data from the vehicle to the central node through the intermediate mains powered by the stationary BLE nodes.The centralized unit consolidates the data analyses, processes, and stores in a prescribed format, providing a detailed view to the industry maintenance team and to the fl eet management team.Either of fl eet management or industry maintenance team can suggest a recommendation in speed control to the vehicle through the intermediate BLE nodes from the centralized server.
The hardware system depicted in Figure 4 receives the data from the sensors to the microcontroller and intelligent embedded algorithm on the IoT edge device to recommend the speed based on the observed values and its priorities as explained in the Flow Chart in Figure 5.The speed correction acts through the actuators based on recommendation obtained either from vehicle internal intelligent micro-controller or from the

SIMULATION AND TESTING
The testing follows a split sub-system architecture type unit test strategy, assuming offthe-shelf connections remain stable and interoperable.It uses appropriate signal generators to simulate the sensors output for evaluating the effi ciency of the fi rmware with a view to control speed engaging a PWM Out signal, read back as the vehicle speed through feedback.It monitors the vehicles Payload, tire, SoC, SoH and Navigation Info as serial data, uses a 2-stage relay for fatigue detection, signal generators for speed, potentiometer for temperatures, and current source for the motor current consumed.The simulation and testing block diagram are shown in Figure 6.
RTOS prioritizes the parameters in the sense it off ers the highest priority to battery related parameters, SoC & SoH, and triggers speed control only if battery exists in the operating condition within limits and the lowest priorities to the truck payload and tire pressure assuming those remain within the limits.The procedure creates a change of vehicles parameters based on the priority for directing the reduced speed to be set as the operating value.Table 3 shows the prioritized parameters.

RESULTS AND DISCUSSION
The process examines the effi ciency of the speed control algorithm for the battery operated off -road industrial electric vehicle against the permissible operating ranges of SoC, SoH, payload It monitors the vehicle parameters in relation to the simulated sensor inputs and tests the vehicle against the recommended speed control through appropriate feedback mechanism.Figure 7 represents the relationship between the input PWM and its output response capturing the vehicle higher battery life and better mileage of the vehicle when operated up to 80% of its maximum speed.
The speed variations of motor in Figure 8 record that the motor speed remains around an optimum of 70% to 80% of its maximum value and bring in a speed reduction when the motor speed attempts to go above 80%.The graph shown in Figure 9 explains restricting the motor to operate around an ideal temperature range of 70 °C for sustaining the longer battery lifespan with maximum power output.The algorithm recommends lowering the load to reduce the vehicle speed within a minimal deviation of (±) 7% and results in reducing the motor temperature when it exceeds 70 °C.
Figure 10 represents the occurrence of fatigue as a Boolean value to indicate the safe operating conditions and the approach entails detecting fatigue only if it arises three or more consecutive times, considering the safety of the workers within the industry, irrespective of the vehicles speed in which the operator receives a steering system  vibrational warning to reduce the vehicle speed to 15% of current speed.On successive fatigue detection, it enables the reduction in speed by additional 15% every time and the vehicle can be bought to zero acceleration gradually.With ignition on, the cycle becomes reset and the speed reduction due to fatigue detection starts over again, assuming the operator enjoys sufficient time to refresh and return to the normal operating condition.
The exercise further investigates the speed control based on the real time GPS inputs monitoring if the vehicle operates within its prescribed predefined boundaries.On any instance of the vehicle crossing over the prescribed boundary, it reduces the speed of the vehicle to 80% of the maximum speed.Besides for every instance of the vehicle crossing the boundary, it lowers the speed immediately reduced by an additional 10%.Figure 11 shows the Speed control tracking vehicle using GPS.
The IoT based intelligent microcontroller monitors the vehicle parameters and takes a combinatorial decision driven by priorities.The results detail the various test cases and facilitate the fact that the algorithm reduces the vehicle speed automatically to the recommended level with 7% deviation and 93% accuracy ensuring safe vehicle operation and guaranteeing longevity to the life of the battery.

CONCLUSIONS
A BLE Mesh network based IoT speed control mechanism was designed for a battery operated industrial off-road electric vehicle suitable for a confined predefined boundary.The architecture was designed on an ARM core based STM32 microcontroller through which it recommends a pulse width variation to switch the converter interface for controlling the BLDC motor responsible for driving the vehicle.The methodology was implemented with the real time input from GPS location services to decide on the set points for controlling the speed.The coding was automated to operate the test bench in accordance with the design of the algorithm and thereby emulate the changing dynamic time bound behaviour of the vehicle.The simulation results were presented to claim the suitability of the scheme for adoption by industries to be integrated into live production lines.
(Fabian et al., 2014; Reif et al., 2014; Wang et al., 2018).However, the communication abilities of the vehicles, irrespective of their off -or on-road nature have been demanding a higher thrust(Schöner2004; Lajunen et al., 2016; Vellucci et al., 2012; Porsche Engineering, 2011; Al-Othman et al., 2016; Club Technical, 2020).It has been projected that the autonomous vehicles avoid many accidents simply by following the traffi c rules and attract the development of new services (Iora and Tribioli 2019; Indira Gandhi National Open University; Mohammad et al., 2016; Lajunen et al., 2012; Liukkonen et al., 2013; Jo et al., 2011).The industrial electric vehicle has been experiencing huge challenges for making fossil fuel-based vehicles autonomous and intelligent.Considerable eff orts have been the order of the day in the growth of the offroad industrial electric vehicle segment and address the challenges that arise for various reasons (Paraszczak et al., 2014; Molet al., 2010; https:// amperevehicles.com/industrial-vehicles/;M N et al., 2018; Lajunen et al., 2016).

Figure 2
Figure 2 refl ects a modular approach to control the speed of the chosen industrial electric goods vehicles with Hardware, Hardware Abstraction Layer, Basic Vehicle Application Layer, Cab Operator Application Layer, IoT Monitoring and Control Interface.The Hardware abstracts the sensor data, validates, analyses, processes by scaling and provides it to the vehicle layer.The software-architecture is shown in Figure 3.STM32F746NG communicates with all sensors in the vehicle and collects the data.The IoT monitoring and control interface layer incorporates the BLE protocol to communicate between the other BLE nodes installed across the premises.The stationary BLE nodes act as repeaters and extenders, carry the vehicle data from the vehicle to the centralized data collection master and there from returns the control information back to the vehicle.The smart edge device powered by STM-32F746NG inherits inbuilt intelligence to control the speed of the vehicle based on the dynamic behaviour of the vehicle and its parameters.The process enables the interface of the Bluetooth Module HC-05 with the STM on SPI (Serial Peripheral Interface) and includes BLE 4.0 stack in it with additional application layer implementation.The stationary BLE nodes installed in the industry can be an off -the-shelf certifi ed interoperable customizable repeater to facilitate communication with the centralized on-premises data collection unit.It identifi es the centralized on-premises unit by a static addressing mechanism to transmit the data from the vehicle to the central node through

Table 1 .
Vehicle specifications

Table 4 .
Operating range of vehicle parameters