Integrating Actual Decision-Making Requirements for Intelligent Collision Avoidance Strategy in Multi-Ship Encounter Situations

COLREGs-abiding hybrid collision avoidance algorithm based on deep reinforcement learning for USVs

Unmanned Aerial Vehicle (UAV) has received wide interest in recent years because of their promising potentials for several applications that include border surveillance, intrusion, detection, emergent item delivery, wildlife monitoring, rescue operations etc. The growing number of entrances of drones into airspace has emerged a significant interest in the collision and obstacle avoidance in autonomous operation. One of major challenges for safer deployment of UAVs is collision avoidance in order to make UAVs autonomous and self-sustainable. In today’s technological era, it is possible for UAVs to avoid obstacles and prevent collisions while navigation by implementing cutting edge computer vision algorithms and visual tracking systems. But it is challenging due to the limited payload in UAV that only permits vision based sensors such as camera, inertial measurement unit such for control and obstacle detection. This paper aims to summarize existing solutions of collision avoidance strategies used in the navigation of UAV. Collision avoidance strategies based on geometry, optimization, sense and avoid, vision-based and force field based methods are investigated in this study.

Aims: To enhance the safety of intelligent vehicle drivers and help them achieve active collision avoidance. Background: Existing research on intelligent vehicles’ active collision avoidance focuses on reference path optimization and stability tracking control. However, there is relatively little research on collision avoidance in emergency accident scenarios. Longitudinal emergency braking collision avoidance technology has been relatively mature, while steering collision avoidance still needs a long time to develop. The research on joint steering and braking collision avoidance in emergency situations is still limited. Objective: Modeling the longitudinal and lateral safety distances based on the safety conditions of adjacent lanes, two active collision avoidance strategies are designed, namely the steering strategy and combined lateral and longitudinal collision avoidance strategy, with the aim of reducing the risk of collision. Methods: In the steering strategy, five control points are first determined sequentially according to steering collision constraints. Then, a collision avoidance path is established based on the quartic Bezier curve. To enhance the tracking of this path, an LQR path-tracking controller is employed to reduce errors via feed-forward and feedback control. In the case of a combined lateral and longitudinal collision avoidance strategy, the LQR and PID controllers are first adjusted and optimized, and then the collision avoidance path is replanned. Besides, the safety risks associated with braking and steering are taken into consideration, and constraints are set on the braking deceleration accordingly. Results: The effectiveness of the designed steering strategy and combined lateral and longitudinal collision avoidance strategy is verified through simulation results obtained by using CarSIM and Matlab/Simulink. They can effectively avoid collisions in emergency situations. Conclusion: The proposed collision avoidance algorithm can avoid collision in emergency scenarios and enhance the safety of autonomous driving.

Real-time motion planning in an unknown environment involves collision avoidance of static as well as moving agents. Strategies suitable for navigation in a stationary environment cannot be translated as strategies per se for dynamic environments. In a purely stationary environment all that the sensor can detect can only be a static object is assumed implicitly. In a mixed environment such an assumption is no longer valid. For efficient collision avoidance identification of the attribute of the detected object as static or dynamic is probably inevitable. Presented here are two novel schemes for perceiving the presence of dynamic objects in the robot's neighborhood. One of them, called the Model-Based Approach (MBA) detects motion by observing changes in the features of the environment represented on a map. The other CBA (cluster-based approach) partitions the contents of the environment into clusters representative of the objects. Inspecting the characteristics of the partitioned clusters reveals the presence of dynamic agents. The extracted dynamic objects are tracked in consequent samples of the environment through a straightforward nearest neighbor rule based on the Euclidean metric. A distributed fuzzy controller avoids the tracked dynamic objects through direction and velocity control of the mobile robot. The collision avoidance scheme is extended to overcome multiple dynamic objects through a priority based averaging technique (PBA). Indicating the need for additional rules apart from the PBA to overcome conflicting decisions while tackling multiple dynamic objects can be considered as another contribution of this effort. The method has been tested through simulations by navigating a sensor-based mobile robot amidst multiple dynamic objects and its efficacy established.

BackgroundCollision avoidance between two walkers involves a mutual adaptation to speed and orientation in order to successfully avoid a collision. Minimum Predicted Distance (MPD) is the distance at which two walkers would collide if their speed and path trajectory were maintained at first sight of one another. MPD has been used to describe the risk of collision and its evolution over time between two adult walkers when on a collision course. Middle-aged children have been shown to have poor perception-action coupling during static and dynamic collision avoidance tasks. Research has yet to examine whether perception-action coupling deficits persist in a dynamic collision avoidance task involving a child and another walker. Research questionCan the metric MPD(t) be used to examine collision avoidance strategies between children and adults? MethodsEighteen children (age: 10 ± 1.5 years) and eighteen adults (34 ± 9.6 years) walked along a 12.6 m pathway while avoiding another participant (child or adult). Groups of three children and three adults were recruited per session. Trials were randomized equally such that each adult interacted with another adult 20 times, each child interacted with another child 20 times, and each adult interacted with a child 21 times, for a total of 141 trials. 3D kinematic data of each participant’s head was recorded using the Vicon system. ResultsThe results demonstrated: (1) MPD(t) can be used to predict future collisions in children, (2) MPD(t) is an absolute measure that is consistently lower when a child is involved compared to two adult walkers, (3) the individual passing second, even when it is a child, contributes more to MPD(t) than the walker passing first. SignificanceIt appears children have developed adult-like strategies during a collision avoidance task involving two walkers. Body anthropometrics should be considered when determining collision avoidance strategies between children and adults.

In order to make the target ships in the Maritime Police Warships Simulation System to avoidance collision automatically, a ship intelligent collision avoidance scheme has been proposed. First, a mathematical model for the determination of the ship's collision risk is advanced based on fuzzy theory. Secondly, a method for the analysis of the Intelligent Collision Avoidance ship and target ship collision risk situation is proposed. Finally, a collision avoidance command decision-making methods is presented. How these intelligent collision avoidance methods are applied to Maritime Police Warships Simulation System is also analyzed.

Underwater collision avoidance using undulating elongated fin propulsion

A formulation that makes possible the integration of collision prediction and avoidance stages for mobile robots moving in general terrains containing moving obstacles is presented. A dynamic model of the mobile robot and the dynamic constraints are derived. Collision avoidance is guaranteed if the distance between the robot and a moving obstacle is nonzero. A nominal trajectory is assumed to be known from off-line planning. The main idea is to change the velocity along the nominal trajectory so that collisions are avoided. A feedback control is developed and local asymptotic stability is proved if the velocity of the moving obstacle is bounded. Simulation results verify the value of the proposed strategy. >

An integrated collision prediction and avoidance scheme for mobile robots in non-stationary environments

A new hierarchical collision avoidance scheme is proposed to coordinate multiple robots in a common workspace by combining the techniques of intelligent control and neural networks (NNs). The high level in the hierarchy is formed by a knowledge-based coordinator (KBC) and an NN-based predictor, and the low level consists of the robots to be coordinated. The authors state the problem of coordinating multiple robots for collision avoidance and the basic principles of the KBC. The knowledge acquisition and representation of collision detection and avoidance for both cylindrical- and revolute-type robots are discussed. Design of the NN-based predictor and the KBC is summarized. The proposed scheme was tested extensively via simulations for both types of robots, showing promising performance. >

With the development of intelligent waterborne transportation, mining collision avoidance patterns based on spatiotemporal and motion data of ships are crucial for the autonomous navigation of intelligent ships, which requires accurate collision avoidance information under various encounter scenarios. Addressing the existing issues of low precision and false detection in data mining algorithms, this paper proposes a collision avoidance behavior mining model considering encounter scenarios. The model is based on the Automatic Identification System (AIS) and the International Regulations for Preventing Collisions at Sea (COLREGs); it firstly identifies ship collision avoidance turning points by analyzing trajectory curvature with turning and recovering factors. Then, by combining AIS data and the specific navigational environment, it matches the ship encounter pairs and determines the encounter scenarios. Comparative experiments show that the model demonstrates superior accuracy in various scenarios compared to traditional algorithm. Finally, the model was applied to AIS data east of the Yangtze River Estuary, recognizing a total of 827 instances of ship collision avoidance behavior under different encounter scenarios. The case study shows that the model can precisely mine collision avoidance information, laying a solid foundation for future research on autonomous collision avoidance decision making for intelligent ships.

When sailing on cluttered water, effective collision avoidance is an essential safety requirement for unmanned surface vehicles (USVs). So that it can avoid multiple random ships effectively. This paper proposes a collision avoidance strategy for USVs based on the probabilistic game theory framework. First, a probability model is established combined with the USV’s kinematics for the collision avoidance problem. Then a greedy algorithm is designed under the framework of probability. It uses a cost function that searches possible movement paths to determine avoidance strategies for the USV. Compared with other collision avoidance methods, the method in this paper seeks to minimize the collision probability instead of guaranteeing no collision, because it is very difficult or impossible to completely avoid the collision in practice. Based on this, our method copes with kinematic constraints of the USV. The effectiveness of the algorithm is verified by simulation, and the simulation results show that the collision avoidance strategy proposed in this paper achieves a very good collision avoidance performance.

Aiming at the problem of ship collision avoidance, an assistant decision-making method based on genetic algorithm to give a better collision avoidance strategy is proposed. The algorithm adopts steering angle, steering time, navigation restoration angle, navigation restoration time as genetic factors. The collision avoidance scheme given by this gene coding method gives specific operation and is more consistent with the practice of navigation. At the same time, the global optimization ability of genetic algorithm is improved by multi group co evolution, adaptive selection operator and Metropolis criteria screening new individuals. According to the requirements in the rules, the initial population is generated randomly in the feasible region of the variable. At the same time, the new individuals generated by the crossover operator and the mutation operator pass the test of the rule, and the new individuals that do not conform to the rule are rejected. The simulation results show that the optimal individual obtained by the improved GA not only satisfies the security requirements, but also has low collision avoidance cost, which is in line with the economic requirements.

This paper presents a viable approach for incorporating collision avoidance strategies into existing guidance and control systems on marine vessels. We propose a method that facilitates the use of simulation-based Model Predictive Control (MPC) for collision avoidance (COLAV) on marine vessels. Any COLAV strategy to be applied in real traffic must adhere to the international regulations for preventing collisions at sea (COLREGS). The proposed MPC COLAV method does not rely on an accurate model of the guidance system to achieve vessel behaviors that are compliant with the COLREGS. Rather, it depends on transitional costs in the MPC objective for collision avoidance maneuvers that are being executed by the marine vessel. Hence, it is straightforward to implement the MPC COLAV on different vessels without specific knowledge of the vessel's guidance strategy. Moreover, it offers the possibility to switch between different (possibly application specific) guidance strategies on the same vessel while running the same MPC COLAV algorithm. We present results from full scale experiments that show the viability of our method in different collision avoidance scenarios.

Computer-controlled, human-like virtual agents (VAs), are often embedded into immersive virtual environments (IVEs) in order to enliven a scene or to assist users. Certain constraints need to be fulfilled, e.g., a collision avoidance strategy allowing users to maintain their personal space. Violating this flexible protective zone causes discomfort in real-world situations and in IVEs. However, no studies on collision avoidance for small-scale IVEs have been conducted yet. Our goal is to close this gap by presenting the results of a controlled user study in a CAVE. 27 participants were immersed in a small-scale office with the task of reaching the office door. Their way was blocked either by a male or female VA, representing their co-worker. The VA showed different behavioral patterns regarding gaze and locomotion. Our results indicate that participants preferred collaborative collision avoidance: they expect the VA to step aside in order to get more space to pass while being willing to adapt their own walking paths.