Collision Avoidance Scheme Research Articles

Modeling collision avoidance maneuvers for micromobility vehicles

Introduction: In recent years, as novel micromobility vehicles (MMVs) have hit the market and rapidly gained popularity, new challenges in road safety have also arisen. There is an urgent need for validated models that comprehensively describe the behavior of such novel MMVs. This study aims to compare the longitudinal and lateral control of bicycles and e-scooters in a collision-avoidance scenario from a top-down perspective, and to propose appropriate quantitative models for parameterizing and predicting the trajectories of the avoidance—braking and steering—maneuvers. Method: We compared a large e-scooter and a light e-scooter with a bicycle (in assisted and non-assisted modes) in field trials to determine whether these new vehicles have different maneuverability constraints when avoiding a rear-end collision by braking and/or steering. Results: Braking performance in terms of deceleration and jerk varies among the different types of vehicles; specifically, e-scooters are not as effective at braking as bicycles, but the large e-scooter demonstrated better braking performance than the light one. No statistically significant difference was observed in the steering performance of the vehicles. Bicycles were perceived as more stable, maneuverable, and safe than e-scooters. The study also presents arctangent kinematic models for braking and steering, which demonstrate better accuracy and informativeness than linear models. Conclusions: This study demonstrates that the new micromobility solutions have some maneuverability characteristics that differ significantly from those of bicycles, and even within their own kind. Steering could be a more efficient collision-avoidance strategy for MMVs than braking under certain circumstances, such as in a rear-end collision. More complicated modeling for MMV kinematics can be beneficial but needs validation. Practical Applications: The proposed arctangent models could be used in new advanced driving assistance systems to prevent crashes between cars and MMV users. Micromobility safety could be improved by educating MMV riders to adapt their behavior accordingly. Further, knowledge about the differences in maneuverability between e-scooters and bicycles could inform infrastructure design, and traffic regulations.

A Review of Artificial Intelligence-Based Optimization Applications in Traditional Active Maritime Collision Avoidance

The probability of collisions at sea has increased in recent years. Furthermore, passive collision avoidance has some disadvantages, such as low economic efficiency, while active collision avoidance techniques have some limitations. As a result of the advancement of computer technology, active collision avoidance techniques have also been optimized by using artificial intelligence-based methods. The purpose of this paper is to further the development of the field. After reviewing some passive collision avoidance schemes, the paper discusses the potential of active obstacle avoidance techniques. A time-tracing approach is used to review the evolution of active obstacle avoidance techniques, followed by a review of the main traditional active obstacle avoidance techniques. In this paper, different artificial intelligence algorithms are reviewed and analyzed. As a result of the analysis and discussion in this paper, some limitations in this field are identified. In addition, there are some suggestions and outlooks for addressing those limitations. In a way, the paper can serve as a guide for the development of the field.

Marine Collision Avoidance Route Planning Model for MASS Based on Domain-Based Predicted Area of Danger

When the own ship encounters target ships in a close-quarter situation, an officer on watch needs to safely and timely alter the course of the vessel to avoid a collision. If ECDIS can automatically collect the navigation parameters and plot areas of collision as quasi-static obstruction areas, it will be much easier for seafarers to implement effective route planning. Hence, this study focuses on developing the MCARP model as a theoretical concept based on the DPAD model and LCD model. By operating the MCARP using ArcGIS, DPADs and several effective collision avoidance routes can be generated and imported into ECDIS based on AIS information at large scales and ample time. The graphic overlay of DPADs and effective routes on ECDIS can serve as a collision avoidance strategy reference for the personnel controlling maritime autonomous surface ships. Finally, different ships encountering situations were input into a Transas navigational simulator. The simulation results showed that own ship could avoid collisions with multiple target ships at distances larger than the preset collision avoidance distances, which also indicated that MCARP is practically feasible.

An adaptive collision avoidance strategy for autonomous vehicle under various road friction and speed

When the autonomous vehicle (AV) is under various road friction and speed, emergency collision avoidance is extremely difficult. In this situation, the AV may encounter severe understeering problem, so it is hard to achieve collision avoidance, even under the control of active safety system. To tackle this problem, an adaptive collision avoidance strategy (ACAS) is proposed for AV under various road friction and speed. The adaptive performance of the ACAS is realized via three aspects. (1) An adaptive reference path planning method is proposed to provide desired evasive speed and reference path for the AV according to various road friction and reduces the turning burden of AV. (2) A predictive-based fuzzy controller is designed to realize the speed control, and it improves the tracking accuracy of various desired evasive speed. (3) A novel turning enhanced method built with a direct yaw turning controller and a torque distribution method can enhance the turning capability of AV. Finally, the proposed strategy is verified on AV via simulation experiments. The code can be found online here: https://github.com/wangjinlei-hnu/ACAS.

Study on Lane-Change Replanning and Trajectory Tracking for Intelligent Vehicles Based on Model Predictive Control

When an intelligent vehicle changes lanes, the state of other vehicles may change, which increases the risk of collision. Therefore, real-time local path replanning is needed at this time. Based on model predictive control (MPC), a lane-change trajectory replanning strategy was proposed, which was divided into a lane-change trajectory correction strategy, a lane-change switchback strategy and forward active collision avoidance strategy according to collision risk. Based on the collision risk function of the rectangular safety neighborhood, the objective functions were designed according to the specific requirements of different strategies. The vehicle lateral controller based on MPC and the vehicle longitudinal motion controller were established. The longitudinal velocity was taken as the joint point to establish the lateral and longitudinal integrated controller. The trajectory planning module, trajectory replanning module and trajectory tracking module were integrated in layers, and the three trajectory replanning strategies of lane-change trajectory correction, lane-change switchback and forward active collision avoidance were respectively simulated and verified. The simulation results showed the trajectory replanning strategy achieves collision avoidance under different scenarios and ensures the vehicle’s driving stability. The trajectory tracking layer achieves accurate tracking of the conventional lane-change trajectory and has good driving stability and comfort.

An efficient genetic algorithm for multi AGV scheduling problem about intelligent warehouse

PurposeThis study aims to address the challenge of automatic guided vehicle (AGV) scheduling for parcel storage and retrieval in an intelligent warehouse.Design/methodology/approachThis study presents a scheduling solution that aims to minimize the maximum completion time for the AGV scheduling problem in an intelligent warehouse. First, a mixed-integer linear programming model is established, followed by the proposal of a novel genetic algorithm to solve the scheduling problem of multiple AGVs. The improved algorithm includes operations such as the initial population optimization of picking up goods based on the principle of the nearest distance, adaptive crossover operation evolving with iteration, mutation operation of equivalent exchange and an algorithm restart strategy to expand search ability and avoid falling into a local optimal solution. Moreover, the routing rules of AGV are described.FindingsBy conducting a series of comparative experiments based on the actual package flow situation of an intelligent warehouse, the results demonstrate that the proposed genetic algorithm in this study outperforms existing algorithms, and can produce better solutions for the AGV scheduling problem.Originality/valueThis paper optimizes the different iterative steps of the genetic algorithm and designs an improved genetic algorithm, which is more suitable for solving the AGV scheduling problem in the warehouse. In addition, a path collision avoidance strategy that matches the algorithm is proposed, making this research more applicable to real-world scheduling environments.

Autonomous Navigation Decision-Making Method for a Smart Marine Surface Vessel Based on an Improved Soft Actor–Critic Algorithm

In this study, an intelligent hybrid algorithm based on deep-reinforcement learning (DRL) is proposed to achieve autonomous navigation and intelligent collision avoidance for a smart autonomous marine surface vessel (SMASV). First, the kinematic model of the SMASV is used, and clauses 13 to 17 of the Convention on the International Regulations for Preventing Collisions at Sea (COLREGs) are introduced. Then, the electronic chart is rasterized and used for path planning. Next, states, actions, and reward functions are designed, and collision avoidance strategies are formulated. In addition, a temperature factor and a constrained loss function are used to improve the soft actor–critic (SAC) algorithm. This improvement reduces the challenges of hyperparameter adjustment and improves sampling efficiency. By comparing the improved SAC algorithm with other deep-reinforcement learning (DRL) algorithms based on strategy learning, it is proved that the improved SAC algorithm converges faster than the other algorithms. During the experiment, some unknown obstacles are added to the simulation environment to verify the collision-avoidance ability of the trained SMASV. Moreover, eight sea areas are randomly selected to verify the generalization ability of the intelligent-navigation system. The results show that the proposed method can plan a path for the SMASV accurately and effectively, and the SMASV decision-making behavior in the collision-avoidance process conforms to the COLREGs in both unknown and dynamic environments.

Automatic Collision Avoidance Route Planning Method in Complex Navigation Environments

Efficient collision avoidance (CA) route planning is one of the important technologies for ensuring the safety of autonomous ships. Many CA route planning studies have been developed for the open sea, but few studies have been conducted for ships in complex navigation environments. A complex navigation environment refers to water encompassing various factors, such as multiple obstacles, varying water depths, ships collision risk, and other maritime challenges. These factors collectively contribute to the complexity of the environment. CA planning in this type of environment is a special type of route planning. To address CA in complex navigation environments, firstly, the route binary tree algorithm is utilized for conducting global route planning in order to determine the optimal path. Secondly, this paper proposes a dynamic generation of local collision avoidance trajectories along the optimal path, seamlessly integrating global route optimization and local collision avoidance. Finally, a comparative analysis is conducted using classical techniques to verify the effectiveness of the proposed method. Numerical simulation results reveal that when a ship is taken as the object of CA in the open sea, the route generated using the proposed method exhibits a remarkable level of unity with the one generated by the benchmark method of improved Artificial Potential Field (APF), particularly in terms of collision avoidance strategy selection, collision avoidance trends, and the actual trajectory. The former’s computational efficiency is improved by at least 30%. When dealing with CA at sea, with complex navigation obstacles, the proposed method considers static complex navigation obstacles, dynamic ships, and their unpredictable strategies to timeously generate a safe path which considers some key rules from the COLREGS (International Regulations for Preventing Collisions at Sea) (IMO, 1972).

Planning and control of drifting-based collision avoidance strategy under emergency driving conditions

Automated emergency collision avoidance (CA) technology has made tremendous progresses in recent years. However, the existing CA approaches, either braking or steering based, cannot handle well the extremely emergency CA scenarios with limited space and time. This is a research gap to be bridged before the zero-fatality vision of road traffic, namely Vision Zero, can be realized. To further extend the safety limits of vehicles, especially that of automated vehicles in future, we propose a drifting-based CA approach that adopts an aggressive, but controllable, over-steering maneuver. To this end, first the trajectory planning of drifting is formulated and solved as a constrained optimal control problem, which considers vehicle dynamics, actuator limits and collision constraints. Then the knowledge of the three CA approaches’ capability limits is obtained and used for CA decision. Given a planned trajectory, a feedback linearization based controller is further designed to complete the drifting maneuver. Finally, using a real vehicle with by-wire actuators, a series of simulations and road tests are carried out. Results show that our proposed drifting-based algorithm can successfully fulfill agile CA tasks in extreme conditions. With further improvements, this preliminary and conceptual work may contribute as a promising complement to current active safety technologies.

A humanoid collision-avoidance strategy considering a large sideslip angle state

In high-speed collision-avoidance scenarios, collision-avoidance requirements in some extreme situations conflict with vehicle stability requirements. Professional drivers deal with these situations in competitions by drifting. Inspired by this technique, this paper proposed a humanoid collision-avoidance strategy that allows vehicles to perform emergency collision-avoidance tasks with a temporary large sideslip angle state. The proposed strategy imitated the reaction process of human drivers and divided the collision-avoidance process into three stages. The first stage was the collision-avoidance stage, in which the target turned the travel direction rapidly by generating a large heading angular velocity. The second stage was the transitional stage, in which the vehicle states were mitigated on the premise that the heading angle was not reduced until the obstacle and the vehicle did not overlap. In the third stage, the vehicle was guided into the side lane by model predictive control. Throughout this process, the vehicle was allowed to temporarily break the stability constraints. Additionally, this paper proposed a regression region constraint to ensure that the vehicle could return to its stability region. Simulations in a variety of conditions and model-in-loop (MIL) experiments were conducted. According to the simulation results, the proposed control strategy could drive the vehicle out of the stability region to deal with more extreme conditions under a regression region constraint. The collision can be avoided in shorter longitudinal and lateral distances by the proposed control strategy. Besides, the MIL experiment verified the real-time performance of the proposed system.

Vision-Based In-Flight Collision Avoidance Control Based on Background Subtraction Using Embedded System.

The development of high-performance, low-cost unmanned aerial vehicles paired with rapid progress in vision-based perception systems herald a new era of autonomous flight systems with mission-ready capabilities. One of the key features of an autonomous UAV is a robust mid-air collision avoidance strategy. This paper proposes a vision-based in-flight collision avoidance system based on background subtraction using an embedded computing system for unmanned aerial vehicles (UAVs). The pipeline of proposed in-flight collision avoidance system is as follows: (i) subtract dynamic background subtraction to remove it and to detect moving objects, (ii) denoise using morphology and binarization methods, (iii) cluster the moving objects and remove noise blobs, using Euclidean clustering, (iv) distinguish independent objects and track the movement using the Kalman filter, and (v) avoid collision, using the proposed decision-making techniques. This work focuses on the design and the demonstration of a vision-based fast-moving object detection and tracking system with decision-making capabilities to perform evasive maneuvers to replace a high-vision system such as event camera. The novelty of our method lies in the motion-compensating moving object detection framework, which accomplishes the task with background subtraction via a two-dimensional transformation approximation. Clustering and tracking algorithms process detection data to track independent objects, and stereo-camera-based distance estimation is conducted to estimate the three-dimensional trajectory, which is then used during decision-making procedures. The examination of the system is conducted with a test quadrotor UAV, and appropriate algorithm parameters for various requirements are deduced.

Underwater collision avoidance using undulating elongated fin propulsion

Underwater navigation to survey littoral regions or complex environments such as offshore structures, coral reefs or sensitive coastal zones requires a highly maneuverable propulsion mechanism. Conventional propulsors have limited maneuverability, often incapable of providing adequate torques and forces to maneuver in such environments. Elongated ribbon fin propulsion, a bio-inspired propulsion technique, can overcome these shortcomings. In the present work, we used a robotic vessel with elongated fin propulsion to test and implement three collision avoidance strategies based on potential field methods. We found that having a circular sensing envelope in the robotic vessel helps to smoothly transition to the mission path after collision is avoided. However, under this strategy, the distance to return to the desired path tends to be longer. The selection of adequate collision avoidance strategies will depend on multiple factors, including others sensing capabilities, desired reaction time, rate of turning and objects to avoid. Although this work focused on relatively simple scenarios for obstacle avoidance, we envision that utilizing more complex maneuver techniques, including rapid turns and hovering, collision avoidance strategies for underwater vessels with undulating fin propulsion can take advantage of the full fin kinematics manipulation to maneuvers as elegant as seen on the fish counterpart.

Research on the strategies for collision avoidance of multi‐UAV with three dimensional formation in combination of consensus algorithm and uniform flow

AbstractThe paper proposes the method of uniform flow loading the improved consensus algorithm to achieve obstacle avoidance for multi‐UAV with 3‐D formation. In view of that the flow function is generally only applicable to two dimensional flow field, dimension reduction is to treat multi‐UAV formation as each UAV flows around along the circular surface of the projected cylinder of spherical obstacle at its different heights. To formation guaranteeing the smooth realization of obstacle avoidance of multi‐UAV formation centered on the leader, it is used the tactics of virtually expanding the radius of these static or moving equivalent cylindrical obstacles to reserve a certain space. The innovation of this strategy is to further reduce the possibility of collision between UAV formation and the edge of obstacles, meanwhile the problem of multi‐UAV formation avoiding obstacles is transformed into the problem of collision avoidance navigation for the leader treated as a particle. It is considered collision avoidance for multi‐UAV formation under three scenarios assumed are the most common situations, and the simulation experimental results of these scenarios are demonstrated on the validity of the designed experiments based on these algorithms.

MARCOL: A Maritime Collision Avoidance Decision-Making Testbed

Safe and efficient maritime navigation is fundamental for autonomous surface vehicles to support many applications in the blue economy, including cargo transportation that covers 90% of the global marine industry. We developed MARCOL, a collision avoidance decision-making framework that provides safe, efficient, and explainable collision avoidance strategies and that allows for repeated experiments under diverse high-traffic scenarios.

Collision-Avoiding Model Predictive Rendezvous Strategy to Tumbling Launcher Stages

This paper considers the situation where a small satellite shall autonomously rendezvous with a tumbling object in a circular low Earth orbit (LEO) and derives a path-based model predictive controller that uses the docking point state and position of the chaser to guide it to a safe docking autonomously. The strategy embeds collision avoidance elements and reduces the computational effort for calculating the pulses to be provided by the thrusters through opportune algebraic manipulations, a Runge–Kutta 4 propagation method using linearized state transition matrices, and implicit embedding of dynamically equivalent thrust models, leading to constant state propagation matrices. Furthermore, the inputs design optimization problem and the embedded collision avoidance scheme are modeled and explicitly crafted as convex problems, contributing positively to low computational requirements. The docking and collision avoidance capabilities of the proposed scheme are extensively tested in an environment that accounts for all the perturbations relevant to LEO frameworks, for realistic thrust schemes, and for uncertainties in the measurement. Numerical results assess which tumbling objects can be docked or not by means of the proposed schemes as a function of the tumbling rates versus the thrust capabilities and hardware uncertainty of the docker.

Global-local hierarchical path planning scheme for unmanned surface vehicles under dynamically unforeseen environments

Unmanned surface vehicles (USVs) should be allowed to respond to dynamic ocean environments and self-adjust their paths safely and efficiently. In this paper, considering the challenges posed by time-varying, partially unknown complex environments, a novel hierarchical motion planning framework is elaborately designed for a USV, which includes a global trajectory optimization and a local reactive collision avoidance strategy. By encapsulating the intricate nature of ocean environment, a global optimization path planning problem is developed to systematically strengthen the model's adaptability to the complex engineering problem. Incorporating adaptive elite selection and fuzzy probability set, an adaptive-elite GA with fuzzy inference (AEGAfi) is devised to fully exploit the underlying optimization problem, providing high-quality global paths. By applying virtual sensory vector onto the USV's sensing module, the COLREG-compliant local-reaction is achieved by governing feasible actions of USVs under dynamically unforeseen environments. Seamlessly bridged by the transition Clothoid path, the linkage between global optimization and dynamic-avoidance is strengthened by softening the replanning time restriction and maintaining path continuity. Eventually, the motion planning framework merits autonomous global-planning and local-reaction in an organically modular manner. Comprehensive simulations and comparisons in various ocean scenarios demonstrate the effectiveness and superiority of the proposed framework.

Distributed dynamic matrix control with constrained optimization for collision and obstacle avoidance of simulated multiple quadcopters

AbstractA system of fast moving quadcopters has a high risk of collisions with neighboring quadcopters or obstacles. The objective of this work is to develop a control strategy for collision and obstacle avoidance of multiple quadcopters. In this paper, the problem of distributed dynamic matrix control (DMC) for collision avoidance among a team of multiple quadcopters attempting to reach consensus in the horizontal plane and yaw direction ( , and ) is investigated. Violations of a predetermined safety radius generates output constraints on the DMC optimization function, which has not been dealt with in the literature. Different from past works, the proposed strategy can perform collision avoidance in the , , and ‐directions. In addition, logarithmic barrier functions are implemented as input rate constraints on the control actions. Extensive simulation studies for a team of quadcopters illustrate promising results of the proposed control strategy and case variations. In addition, DMC parameter effects on the system performance are studied, and a successful study for obstacle avoidance is presented.

Collision Avoidance Strategy for Unmanned Surface Vessel Considering Actuator Faults Using Kinodynamic Rapidly Exploring Random Tree-Smart and Radial Basis Function Neural Network-Based Model Predictive Control

Path planning and tracking are essential technologies for unmanned surface vessels (USVs). The kinodynamic constraints and actuator faults, however, bring difficulties in finding feasible paths and control efforts. This paper proposes a collision avoidance strategy for USV by developing the kinodynamic rapidly exploring random tree-smart (kinodynamic RRT*-smart) algorithm and the fault-tolerant control method. By utilizing the triangular inequality and the intelligent biased sampling strategy, the kinodynamic RRT*-smart shows its advantages in terms of path length, cost and running time. With consideration of kinodynamic constraints, a feasible and collision-free trajectory can be provided. Then, a radial basis function neural network-based model predictive control (RBF-MPC) method was designed that compensates for the model’s uncertainties by developing the radial basis function neural network (RBF-NN) approximator and by constructing a feedback-state training dataset in real time. Furthermore, two types of fault situation were analyzed considering the thruster failure. We established the faults’ mathematical models and investigated the fault-tolerant strategies for different fault types. The simulation studies were conducted to validate the effectiveness of the proposed strategy. The results show that the proposed planning and control methods can avoid obstacles in faulty conditions.

A comprehensive survey on non‐cooperative collision avoidance for micro aerial vehicles: Sensing and obstacle detection

AbstractIn recent years, unmanned aerial vehicles (UAVs) have been confirmed as a powerful tool for countless applications in nearly every industry, in which collision avoidance plays a vital role for micro aerial vehicles (MAVs) performing autonomous missions. The complexity of collision avoidance systems is directly related to the complexity of the navigation environment, the avoidance of highly uncertain obstacles being a more thoroughly researched critical aspect. The non‐cooperative collision avoidance strategy is necessary when communication between the agents is unavailable, under the case with many existing passive obstacles and non‐cooperative flying objects. In this article, we review the topic of non‐cooperative collision avoidance for MAVs in dynamic environments by covering two main issues: onboard sensing and obstacle description algorithms. A detailed discussion of these two topics is provided to face non‐cooperative collision avoidance technology for MAVs. Finally, the paper summarizes the open challenges and foreseen future directions in this field.

Review on the Real-time Implementation of IoT-enabled UAV in Precision Agriculture and the Overview of Collision Avoidance Strategies

Drone-based monitoring is very convenient and effective in the modern era for monitoring large and dense areas where humans cannot monitor efficiently. Precision agriculture (PA) unmanned aerial vehicle (UAV) monitoring techniques make farmers more protective and knowledgeable about their fields. Modern PA drone employs a 5G-enabled Internet of Things (IoT) that captures field data and transmits it to the cloud with extremely low latency to make quick decisions. From anywhere, farmers can keep an eye on their farms. Also, they have a choice of manual or automated methods for executing the proper data-driven actions. Smart farming is substantially more efficient than traditional farming. This article gives an in-depth analysis of UAVs’ real-time deployment of hardware, software, sensors, and IoT in agriculture for crop monitoring, weed identification, and collision avoidance. This research also covers the wide variety of collision avoidance algorithms utilized in both outdoor and interior conditions.