Robotic Vehicle Research Articles

Monte Carlo tree search with spectral expansion for planning with dynamical systems.

The ability of a robot to plan complex behaviors with real-time computation, rather than adhering to predesigned or offline-learned routines, alleviates the need for specialized algorithms or training for each problem instance. Monte Carlo tree search is a powerful planning algorithm that strategically explores simulated future possibilities, but it requires a discrete problem representation that is irreconcilable with the continuous dynamics of the physical world. We present Spectral Expansion Tree Search (SETS), a real-time, tree-based planner that uses the spectrum of the locally linearized system to construct a low-complexity and approximately equivalent discrete representation of the continuous world. We prove that SETS converges to a bound of the globally optimal solution for continuous, deterministic, and differentiable Markov decision processes, a broad class of problems that includes underactuated nonlinear dynamics, nonconvex reward functions, and unstructured environments. We experimentally validated SETS on drone, spacecraft, and ground vehicle robots and one numerical experiment, each of which is not directly solvable with existing methods. We successfully show that SETS automatically discovers a diverse set of optimal behaviors and motion trajectories in real time.

Cooperative Motion Planning of Multiple Automated Vehicle Robots: A Quick IoT-Based Approach

Cooperative Motion Planning of Multiple Automated Vehicle Robots: A Quick IoT-Based Approach

Development of Software Interface for AI-Driven Weed Control in Robotic Vehicles, with Time-Based Evaluation in Indoor and Field Settings

Herbicide blanket application is widely practiced by farmers to chemically control weeds in the field, thereby enhancing productivity and improving crop quality. However, their repetitive use also has caused several negative issues on the environment and human health. To overcome these negative issues precision site-specific weed control can be used to reduce the herbicide application by targeted application on only weed areas. Hence, a robotic platform utilizing NVIDIA Jetson embedded device as control and processing unit was developed and tested on lab and field conditions for weed control. The average time taken for data acquisition and artificial intelligence-based computer vision tasks was tested under both lab and field conditions. Moreover, a grid map creation algorithm using YOLOv4 deep learning algorithm was evaluated to control the nozzles of the robotic platform. Integrating all these components together enabled real-time weed management approaches, enhancing the precision and efficiency of herbicide application Independent paired t-tests reveal that there is no significant difference in computational time between lab and field testing. Conversely, independent paired t-test revels that there is significant difference in image size between lab and field testing. The reduction of herbicide application based on grid map was obtained 79 to 80 %, which does not include YOLOv4 algorithm failure and system synchronization failure. The study suggests the importance of field testing for real-time applications of the robotic platform by using deep learning computer vision methods for weed control.

Tri-band planar antenna for a small robotic vehicle

Abstract This paper describes the design, analysis of properties and fabrication of a tri-band antenna for a small robotic vehicle. The E-shaped tri-band antenna was chosen for its simplicity, practicality and universality. The design of the tri-band antenna and the simulation of its properties were implemented in the Matlab programming environment using the Antenna Toolbox extension. The designed tri-band antenna was fabricated, its properties were measured and the measured results are compared with the simulations for subsequent analysis.

Fabrication of Remotely Controllable Robotic Weed-Ejector Vehicle

Weed is an unwanted plant in an agriculture field that keeps the crop plants deprived of sunlight, fertilisers and water. If not removed, weed reduces the crop output to a larger extent i.e., endangering the farmer's interests. There are three main reasons for leaving weed to grow in fields and substantial loss of crop output 1) non-availability of agricultural labour, 2) fear of snake bites and 3) higher amounts of time required for weed injection. As of there are no successful ROVs (Remotely operating vehicles) to get to the weed and capable of pulling off the weed without the farmer physically entering into the agricultural fields. The Proposed model reduces farmers getting exposed to snake bites during weed ejection and reduces the time and labour requirement for weed ejection. The project involves the study of different mechanisms required for weed pull-off from the agricultural fields and remote-control systems, modelling of components, fabrication of components, assembly and testing of the Remotely Controllable Robotic Weed-Ejector Vehicle.

Efficient Data Collection Strategy for Modeling the Dynamics of Floating Robotic Vehicles Using Neural Networks

As the hydrodynamic model of a surface vehicle is highly nonlinear and of high dimension, an extensive set of experiments is required to estimate the parameters of the model. Typically, the dynamic behavior of a surface vehicle is investigated through experimental tests such as planar motion mechanisms or free running tests in specialized test facilities. However, the cost of such tests is very high, and it is challenging to design effective and efficient test conditions, especially when the model is represented by numerous parameters such as neural networks. In this paper, we propose an efficient data collection strategy for modeling the dynamics of a floating robotic vehicle using neural networks. For generalized modeling, it is important to collect diverse data by exploring all the reachable state space of the vehicle. To achieve this, a data collection policy is built based on sampling-based planning toward reducing the uncertainty of the neural network-based model. The proposed method allows for collecting more comprehensive data than other commonly used random-based data collection policies. We show experimentally that the proposed method enables more accurate modeling of both fully actuated and under-actuated floating robotic vehicles, as demonstrated by the effective execution of various tasks.

Robust humanoid robot vehicle ingress with a finite state machine integrated with deep reinforcement learning

Robust humanoid robot vehicle ingress with a finite state machine integrated with deep reinforcement learning

Learning to control and coordinate mixed traffic through robot vehicles at complex and unsignalized intersections

Learning to control and coordinate mixed traffic through robot vehicles at complex and unsignalized intersections

Longitudinal robust dynamic programming control for driving robot vehicles with performance self-learning

To realise the accurate and stable longitudinal control of the vehicle manipulated by the driving robot (DRV) under different test conditions, a longitudinal robust dynamic programming (RDP) control method for the DRV based on performance self-learning is proposed. Firstly, the dynamics of the driving robot and the vehicle are analysed. Then, the adjoint sensitivity method is used for the DRV to train the neural ordinary differential equation (NODE). The performance of the vehicle under different pedal openings is learned. A RDP controller is designed to solve the nominal target acceleration. Moreover, the unscented Kalman filtering (UKF) method is used to estimate the uncertain external disturbance of the DRV. Furthermore, dynamic compensation is used for the driving robot to track the vehicle's speed. Finally, the stability of the controller is proved. Compared with MPC, LQR, and PID methods, test results show the effectiveness of the proposed method.

Design of multimodal transformable wheels for amphibious robotic vehicles

Amphibious vehicles are versatile for locomotion on both water and land; to reduce the complexities of structure and control, an amphibious vehicle is expected to use a single propulsion system without additional screw propellers or water jets; as vehicles are typically driven by wheels that are efficient for smooth land locomotion, how to make the wheels also as water propellors for multimodal transformations is meaningful, while such wheels are less explored. From this motivation, in this short article, we report the design of a multimodal transformable wheel with three driving modes (wheel mode, forward-propellor mode, and backward-propellor mode) and the integrated mode-locks for each driving mode. To achieve the multimodal transformations of the wheel, we design the integrated mode-transform mechanism and mode-lock mechanism, the two mechanical implementations are driven by only two actuators and are compact with low inertia and low control complexity, with the kinematic properties provided during the wheel transformations.

English

English

Integrated evaluation of the functioning quality of robotic agricultural vehicle system

The article proposes to use a method for evaluating the effectiveness (Data Envelope Analysis) of a system of robotic agricultural vehicles, which is based on solving the problem of fractional linear programming when analyzing a set of homogeneous production facilities. This method is used for an integrated assessment of the quality of service of robotic vehicles according to the specified key parameters of functioning. The main parameters of the effectiveness of the system of robotic agricultural vehicles are selected: the coefficient of technical readiness, operational speed, downtime in maintenance and repair, the average daily duration of the car, the average monthly mileage of the vehicle, the remaining life of the car, the unit cost of maintenance and specific cargo displacement. Solving systems of linear equations makes it possible to determine the position of each analyzed object relative to the efficiency boundary in a multidimensional parameter space. The solution of the Bunker-Charnes-Cooper problem has been completed. Sets of target changes in the parameters of robotic cars have been obtained, which transfer objects to the efficiency boundary. The analysis of the parameters is carried out by constructing diagrams of production capabilities in a multidimensional space. For two periods of operation of the robotic vehicle system for an agro-industrial enterprise, the Malmquist index was calculated to assess the dynamics of operational efficiency.

Design of Mechatronics Based Virtual Telepresence for Robotic System Using Raspberry Python Interpreter

A telepresence robot is a remote-controlled, wheeled device with a display to enable video streaming which enable the participants to view remote locations, as if they were there. The project consists of a VR headset, with a smartphone in dual screen to experience virtual reality and 4 wheeled robotic vehicles. The movement of the Robot is controlled using a remote controller. The motion of the camera of the robot is controlled by the accelerometer and magnetometer data processed by Arduino and Raspberry Pi. Video streamed is received by the smartphone using the IP address specified by the Raspberry Pi. This robot with a camera is placed in a remote location to capture the environment in visual form using Raspberry Pi (R Pi).The captured visuals are displayed on the user’s virtual reality (VR) headset. An added feature allows the camera to move in the direction of the user’s head movements. This gives the user a real-time experience as if he is present where the robot is located. The robot can also be moved in any direction through an app installed in the user’s smartphone.

Design and Implementation of a Robotic Car with a ULV Atomizer for Mosquito Control in Urban Parks

Mosquito infestations in tropical regions, particularly during months of moderate to high rainfall, pose significant health risks due to diseases like Malaria, Dengue, and Chikungunya. Traditional control methods such as fumigation and chemical sprays face limitations of environmental concerns and potential health risks to humans. To address this issue in community parks, I developed a robotic car equipped with an atomizer to disperse an organic mosquito repellent. Powered by a Raspberry Pi 4 and brushless DC motors, the robotic vehicle is designed for stable and efficient movement across uneven park terrains. It is controlled via an integrated PlayStation 4 controller for user-friendly operation. The atomizer disperses a repellent from essential oils such as Neem (Azadirachta indica), Lemongrass, and Eucalyptus, offering an environmentally friendly and human-safe alternative to chemical repellents like DEET. Testing indicated that the robotic car equipped with an atomizer effectively reduces mosquito presence, enhancing the safety and enjoyment of outdoor recreational spaces.

Field Test of the MiniRadMeter Gamma and Neutron Detector for the EU Project CLEANDEM.

In the framework of the H2020 CLEANDEM project, a small robotic vehicle was equipped with a series of different sensors that were developed for the preliminary inspection of areas possibly contaminated by radiation. Such unmanned inspection allows to identify dangerous locations prior to the possible start of human operations. One of the developed devices, named the MiniRadMeter, is a compact low-cost sensor that performs gamma and neutron radiation field mapping in the environment. The MiniRadMeter was successfully tested in a simulated field mission with four "hidden" radioactive sources and a neutron generator. In this work, we describe the test procedure and the results, which were supported by the outcome of dedicated Monte Carlo simulations.

Experimental assessment of strength characteristics of anthropogenic soils in a given mining area for justification of tractor propulsion type

BACKGROUND: In mining, there is a need to develop tractors as special robotic vehicles of the robotic mining facility that can efficiently move and work on soils with low bearing capacity, in waterlogged areas and in the underwater position. A preliminary evaluation of the strength characteristics of such a heterogeneous surface of the exploitation area helps to choose an effective concept and technical characteristics of the propulsor of the tractor. AIM: Formation of the database of the physical condition and mechanical properties of the environment in which the tractor will move and operate. METHODS: In order to achieve the aim, preliminary experiments were conducted on the terrain. Typical kinds of soil sites and watered areas, including those of anthropogenic origin, of a given mining area were identified. Full-scale instrumental tests were conducted to determine: granulometric composition of soils; physical and mechanical parameters of these soils at the selected sites using the developed original types of a penetrometer and a shear rig. RESULTS: The database of numerical values of measured strength characteristics of one of the soil types (as an example) as a possible supporting surface for vehicle motion in the area of mining facility operation is presented in a graphical form. The basis of this database is the graphs according to the results of experiments on free settlement and shear with the help of dies of two types of configuration (the round die and the original caterpillar track respectively); the granulometric composition of the selected soils has been evaluated. CONCLUSION: The methodology of preliminary assessment of the soil condition by two indicators of soil strength (the rut depth z depending on the average pressure under the die) and shear (average tangential stresses τ in the contact patch on the motion of the die relative to the soil, depending on the ratio of the normal force and the traction force in the contact plane) has been developed. The 3D model of the coastal and bottom area of the technogenic mining zone has been developed. Design variants of highly efficient types of tractor propulsors are selected on the basis of prediction of maximum allowable stresses in the contact patch with the ground for the bearing surface areas of the considered area of operation.

Robot Vehicle with ESP32 CAM

Abstract: A surveillance robot equipped with ESP32-CAM is a setup that assembles a portable surveillance gadget from a robot frame and the ESP32-CAM board. The ESP32-CAM board is an affordable development platform featuring an integrated camera and Wi-Fi capabilities. This robot is capable of navigating and capturing video footage across various settings due to its frame. The gadget can be operated through a web-based interface on the ESP32-CAM board. Through this interface, users can direct the robot's movements, view real-time video feeds, and capture video snapshots. Moreover, the system can be configured to alert the user whenever it detects movement through computer vision methods such as tracking and identifying objects. This ESP32-CAM surveillance robot finds applications in home security, monitoring from afar, and industrial surveillance. Its low cost and user-friendly interface make it an attractive choice for those looking to remotely oversee their surroundings.

Field-based multispecies weed and crop detection using ground robots and advanced YOLO models: A data and model-centric approach

The implementation of a machine-vision system for real-time precision weed management is a crucial step towards the development of smart spraying robotic vehicles. The intelligent machine-vision system, constructed using deep learning object detection models, has the potential to accurately detect weeds and crops from images. Both data-centric and model-centric approaches of deep learning model development offer advantages depending on environment and non-environment factors. To assess the performance of weed detection in real-field conditions for eight crop species, the Yolov8, Yolov9, and customized Yolov9 deep learning models were trained and evaluated using RGB images from four locations (Casselton, Fargo, and Carrington) over a two-year period in the Great Plains region of the U.S. The experiment encompassed eight crop species—dry bean, canola, corn, field pea, flax, lentil, soybean, and sugar beet—and five weed species—horseweed, kochia, redroot pigweed, common ragweed, and water hemp. Six Yolov8 and eight Yolov9 model variants were trained using annotated weed and crop images gathered from four different sites including combined dataset from four sites. The Yolov8 and Yolov9 models’ performance in weed detection were assessed based on mean average precision (mAP50) metrics for five datasets, eight crop species, and five weed species. The results of the weed and crop detection evaluation showed high mAP50 values of 86.2 %. The mAP50 values for individual weed and crop species detection ranged from 80.8 % to 98 %. The results demonstrated that the performance of the model varied by model type (model-centric), location due to environment (data-centric), data size (data-centric), data quality (data-centric), and object size in the image (data-centric). Nevertheless, the Yolov9 customized lightweight model has the potential to play a significant role in building a real time machine-vision-based precision weed management system.

Design and application of a vehicle robot to FAST reflector surface

PurposeThis paper aims to introduce a wheeled vehicle robot for adapting to the surface terrain of the 500-m diameter reflector of the FAST radio telescope in China.Design/methodology/approachBy analyzing vehicles applied for different off-road environments, a six-wheeled architecture with a passive “triple-bogie” suspension is selected. A subscale model of the vehicle robot is designed, along with statics modeling and multibody simulations of the dynamics on simulated reflector panel surfaces. The slope- and step-climbing abilities of the subscale vehicle are discussed in accordance with numerical and experimental tests. An engineering scale vehicle is subsequently manufactured and tested on surface terrains of lateral as well as vertical gaps, and is finally validated on the FAST reflector.FindingsThis model of vehicle robot exhibits strong structure stability under desired payload. It can stably cross lateral gaps for maximum surface slope 28° and can traverse vertical gap for maximum surface slope 23°. The traversing abilities satisfy the mobility requirements subjected to surface terrains of FAST reflector.Originality/valueThe engineering vehicle robot negotiates the lateral as well as vertical gaps between triangle panels and has been successfully applied to the FAST reflector serving for inspection and maintenance work.

Многофакторный анализ эффективности автономных транспортных средств на основе метода анализа среды функционирования

The design of robotic and unmanned vehicles includes a virtual commissioning stage that provides an analysis of the operation of objects using digital twins. Conducting a large number of experiments on models leads to the use of methods for evaluating the results obtained to choose the best design solution. A method of multifactorial anal-ysis of the efficiency of the complex of autonomous vehicles (AV) for agro-industrial purposes is proposed. Multivariate analysis is performed at the virtual commissioning stage in order to plan early maintenance and repair activities. The paper uses the method of analyzing the operating environment for a comparative assessment of various operating scenarios. The method is based on solving a complex of optimization problems of linear programming. A formal description of virtual test scenarios is proposed. The method of multifactorial analysis is implemented in the process of virtual tests using optimal purpose system models and AV simulation. A procedure for virtual AV tests has been developed, including step-by-step modeling and multivariate analysis. The sets of input and output key parameters of the AV and analyzed scenarios are determined. It is proposed to perform two consecutive efficiency assessment tasks. The first task is to compare the effectiveness of individual AV and determine the limits of their effectiveness. The solution of the second task makes it possible to evaluate the effectiveness of AV system operation scenarios. The target values of changes in the key parameters of the AV and operating scenarios are obtained, leading to an increase in the efficiency of operation and maintenance. Conducting a multifactorial analysis of the results of virtual tests allows you to formulate requirements and recommendations for the design of AV maintenance systems.