Collision Avoidance Mechanism Research Articles

SDA: A NOVEL SKEWED-DEEP-ARCHITECTURE FOR VEHICLE MOTION DETECTION IN DRIVING VIDEOS

Collision avoidance mechanisms are important topics for studies in the field of autonomous vehicles. We could obtain prior information about the collision from the movement angles of vehicles. Therefore, it is important issue to learn the movement angles of vehicles in motion. In the study, an architectural model is developed that learns the horizontal movement angles of vehicles to form a base for collision warning systems. YOLOv3 is modified and used on motion profiles. Thanks to the learned angle values, also the bounding boxes match the traces in the motion profiles smoothly. The results obtained have a mAP value of 79% and an operating speed of 36 FPS. These results are better than when trained on motion profiles of the YOLOv3 architecture. In addition, the use of the new architecture on motion profiles and factors such as noise and bad weather in the image do not adversely affect the results. With these features, a fundamental step has been taken for anti-collision systems.

Modelling motorized and non-motorized vehicle conflicts using multiagent inverse reinforcement learning approach

Microsimulation models are effective for analysing road users’ interaction behaviour and assessing different facilities’ performance. However, only a few studies have developed simulation models for studying motorized and non-motorized vehicles conflicts. This is likely due to mixed traffic’s complexity and heterogeneity and the difficulty in accurately capturing road users’ avoidance maneuver. This study aims to adopt a multiagent simulation model to replicate road users’ microscopic behaviour and collision avoidance mechanisms in traffic conflict scenarios. Road users’ reward functions are recovered by the multiagent inverse reinforcement learning approach. The multiagent Actor-Critic deep learning algorithm is used to predict road users’ evasive action and assess their optimal policies. The findings demonstrate that the multiagent simulation model provides highly accurate predictions of road users’ trajectories and collision avoidance strategies. Furthermore, the results demonstrate a strong correlation between the predicted traffic conflict indicator from the simulated trajectories and that from the actual trajectories.

ISAC-Assisted Collision Avoidance Mechanism for Vehicle-to-Infrastructure Systems

ISAC-Assisted Collision Avoidance Mechanism for Vehicle-to-Infrastructure Systems

Predicting pedestrian trajectories at different densities: A multi-criteria empirical analysis

Predicting human trajectories is a challenging task due to the complexity of pedestrian behavior, which is influenced by external factors such as the scene’s topology and interactions with other pedestrians. A special challenge arises from the dependence of the behavior on the density of the scene. In the literature, deep learning algorithms show the best performance in predicting pedestrian trajectories, but so far just for situations with low densities. In this study, we aim to investigate the suitability of these algorithms for high-density scenarios by evaluating them on different error metrics and comparing their accuracy to that of knowledge-based models that have been used since long time in the literature. The findings indicate that deep learning algorithms provide improved trajectory prediction accuracy in the distance metrics for all tested densities. Nevertheless, we observe a significant number of collisions in the predictions, especially in high-density scenarios. This issue arises partly due to the absence of a collision avoidance mechanism within the algorithms and partly because the distance-based collision metric is inadequate for dense situations. To address these limitations, we propose the introduction of a novel continuous collision metric based on pedestrians’ time-to-collision. Subsequently, we outline how this metric can be utilized to enhance the training of the algorithms.

Dynamic adaptive autonomous navigation decision-making method in traffic separation scheme waters: A case study for Chengshanjiao waters

A dynamic adaptive decision-making method in traffic separation scheme (TSS) waters (Chengshanjiao) was explored to achieve fully autonomous navigation. A static traffic environment comprising the digital models of the constituent parts of a TSS was established. A dynamic model was constructed using the target ships. Static and dynamic traffic environments serve as a foundation for decision-making. Based on the mathematical evaluations of the 1972 International Regulations for Preventing Collisions at Sea (COLREGs) and good seamanship, the requirements for motor-vessel collision avoidance in different scenarios and stages were presented. An automatic navigation method was established that could control the ship, automatically track the route, and assess the ship motion state. A collision risk identification model and dynamic collision avoidance mechanism for this type of water were built to identify the most dangerous target ship and a feasible manoeuvring plan to ensure safety in a multi-ship environment. To adapt to the random movement of target ships and correct residual errors, an autonomous navigation decision framework was proposed based on time-series rolling calculations. The ship safely moved through the water by altering its course 5° to starboard at 326 s, 6° to starboard at 887 s, route tracking at 1627 s, altering its course 6° to starboard at 4072 s, and route tracking at 4813 s in an experimental scenario with eight target ships. Six of the eight target ships were randomly manipulated. The results demonstrate that this decision-making technique can be utilised in practice and is precise to 1°.

Maneuver Decision-Making Method for Ship Collision Avoidance in Chengshantou Traffic Separation Scheme Waters

A maneuvering decision-making model based on time series rolling and feedback compensation methods is proposed to solve the problem of high traffic risk in Chengshantou traffic separation scheme (TSS) waters. Firstly, a digital traffic environment model suitable for the TSS waters is proposed. Secondly, a navigation risk identification method in these waters is constructed based on the digitized traffic environment and situation identification model in the Chengshantou TSS waters. Thirdly, considering the requirements of the rules and good seamanship, minimum course altering is obtained by combining the collision avoidance mechanism. Lastly, a maneuvering decision-making model in the TSS waters based on time series rolling and feedback compensation methods is developed. The simulation results show that the ship can correctly identify the collision risk and appropriately obtain maneuvering decisions, and can resume the planned route under the premise of ensuring safety. When the target ships alter course or change speed, the ship can also make adaptive maneuvering decisions. In summary, the proposed method meets the requirement of safe navigation in Chengshantou waters and provides a theoretical basis for the realization of intelligent navigation in waters similar to TSS.

Finite-time adaptive event-triggered control for USV with COLREGS-compliant collision avoidance mechanism

This paper focuses on the real time intelligent navigation scheme for unmanned surface vehicle (USV) subject to collision avoidance regulations. The proposed approach is formulated as two segments: the real time guidance principle with COLREGS-compliant collision avoidance mechanism and the finite-time adaptive event-triggered control algorithm. The pointed improvement for the artificial potential field (APF) can guarantee USV possess the excellent avoidance function in accordance with COLREGS regulations. Especially, when the obstacle makes dangerous actions, the emergency obstacle avoidance operation (EOAO) is implemented according to the real time position of obstacles. Based on the proposed guidance principle, a finite-time robust neural control algorithm is developed by fusion of the finite-time stability theory and event-triggered mechanism. The intermittent computing and communication can effectively reduce the energy and resource consumption of the control system. Moreover, the finite-time adaptive controller design contributes to smaller computation load and facilitates the implementation of the algorithm in ocean engineering. The semi global finite-time uniformly boundedness (SGFTUB) of the closed-loop system is guaranteed through the Lyapunov theory. Finally, the numerical simulation and comparison experiment are illustrated to verify the guidance and control performance of the proposed algorithm.

Collision-Free Distributed Control for Multiple Quadrotors in Cluttered Environments With Static and Dynamic Obstacles

To ensure robust flight of multiple quadrotors in unknown environments, this letter proposes a distributed cooperative control algorithm with a separation-merge mechanism to solve the problem of safe flight and formation reconfiguration. The separation-merge framework integrates the obstacle avoidance mechanism, inter-robot collision avoidance mechanism, and formation reconfiguration mechanism, and couples them all to the control layer. Obstacle detection mechanism and velocity adjustment mechanism are designed based on the obstacle avoidance behavior of the pigeons. And a rotational potential energy is introduced to solve the case of local minima and oscillations where the rotation angle can be changed in real-time according to the relative position. Inspired by spring damping system, a collision avoidance mechanism is designed with the help of Hooke's law with damping. A formation reconfiguration mechanism is designed based on consensus theory, which enables multiple quadrotors to quickly return to the prescribed formation after obstacle avoidance. The proposed method is integrated into a multi-quadrotor system and its effectiveness is verified in the visualization simulator and outdoor experiments.

Reinforcement Learning-based Wi-Fi Contention Window Optimization

The collision avoidance mechanism adopted by the IEEE 802.11 standard is not optimal. The mechanism employs a binary exponential backoff (BEB) algorithm in the medium access control (MAC) layer. Such an algorithm increases the backoff interval whenever a collision is detected to minimize the probability of subsequent collisions. However, the increase of the backoff interval causes degradation of the radio spectrum utilization (i.e., bandwidth wastage). That problem worsens when the network has to manage the channel access to a dense number of stations, leading to a dramatic decrease in network performance. Furthermore, a wrong backoff setting increases the probability of collisions such that the stations experience numerous collisions before achieving the optimal backoff value. Therefore, to mitigate bandwidth wastage and, consequently, maximize the network performance, this work proposes using reinforcement learning (RL) algorithms, namely Deep Q Learning (DQN) and Deep Deterministic Policy Gradient (DDPG), to tackle such an optimization problem. In our proposed approach, we assess two different observation metrics, the average of the normalized level of the transmission queue of all associated stations and the probability of collisions. The overall network’s throughput is defined as the reward. The action is the contention window (CW) value that maximizes throughput while minimizing the number of collisions. As for the simulations, the NS-3 network simulator is used along with a toolkit known as NS3-gym, which integrates a reinforcement-learning (RL) framework into NS-3. The results demonstrate that DQN and DDPG have much better performance than BEB for both static and dynamic scenarios, regardless of the number of stations. Additionally, our results show that observations based on the average of the normalized level of the transmission queues have a slightly better performance than observations based on the collision probability. Moreover, the performance difference with BEB is amplified as the number of stations increases, with DQN and DDPG showing a 45.52% increase in throughput with 50 stations.

Ship collision avoidance based on brain storm optimization near offshore wind farm

An intelligent ship collision avoidance mechanism based on improved swarm intelligence is presented. To expedite optimal decision-making system to prevent vessel collision, an adaptive brain storm optimization (ABSO) with variational inference-based expectation maximum for Mean-shift is proposed. The threshold for random update of individuals in brain storm optimization algorithm was adaptively renewed. The fitness was based on turning amplitude, and distance from original route. The proposed algorithm was tested under different scenarios of ship encounter, including crossing, head-on, and overtaking situations. The rationale is validated via typical cases. One of the minor objectives of the optimization was to prevent ships from steering to the side of the wind farm, which ensures the safety of navigation near offshore wind farm. This algorithm was used in collision avoidance near the offshore wind farm. Results show that calculations based on the proposed optimization method contribute to reasonable and effective decision-making.

Conflict-based search with D* lite algorithm for robot path planning in unknown dynamic environments

This study proposes a locally observable robot pathfinding algorithm, conflict-based search with D* lite (CBS-D*), to realize automatic and effective pathfinding in mixed environments with dynamic obstacles. This algorithm takes Manhattan distance as the heuristic function and extends the incremental search range from 1order to 3order neighbors. It presents a prejudgment mechanism of collision avoidance and investigates a wait and circuity strategy to promote pathfinding performance. Compared with the D* lite, the experimental results demonstrate that CBS-D* achieves a higher success rate and obstacle avoidance number, and a lower time step. By this collision avoidance mechanism, CBS-D* gives all successes in pathfinding in various dynamic environments, while D* lite may result in some failures. Specifically, CBS-D* has around 31% in the average success rate of pathfinding improved to D* lite in a 32 × 32 map. Furthermore, CBS-D* gives a superiority of self-adaptability and intelligence in unknown dynamic environments.

Decentralized modular hybrid supervisory control for the formation of unmanned helicopters

Formation control of Unmanned Aerial Vehicles (UAVs) requires them to tightly cooperate to reach and keep the formation, while avoiding collision. This paper proposes a novel decentralized hybrid supervisory control approach for the formation control of multiple UAVs. This is achieved by developing a symbolic motion planning technique to polarly partition the motion space resulting in a finite state discrete event model for the motion dynamics of each UAV. Then, a modular discrete supervisor is designed for different components of the formation mission including reaching the formation, keeping the formation, and collision avoidance. Further, for the collision avoidance mechanism, a novel top-down decomposition-based approach is developed to design local supervisors decentralizedly. It is formally proved that with the proposed top-down decomposition-based approach, the (locally) supervised UAVs, as a whole, can cooperatively satisfy the desired (global) collision avoidance specification. The proposed decentralized supervisory control algorithm is also verified through a hardware-in-the-loop simulator for the formation control of unmanned helicopters.

Collision-free formation tracking control for multiple quadrotors under switching directed topologies: Theory and experiment

Multiple quadrotors are becoming increasingly attractive in many fields, and the formation tracking control of multiple quadrotors is the foundation for performing complex tasks. In this paper, the collision-free formation tracking control problem for multiple quadrotors under switching topology is studied, and the leader has the control command, which is unknown to the followers. A formation tracking control protocol is firstly constructed utilizing local neighboring information, in which the neighbors of the quadrotors can change. Then, inspired by the spring-damping system, a collision avoidance mechanism based on Hooke's law with damping is proposed to avoid collision of multiple quadrotors in a cluttered flight. By using the Lyapunov based stability analysis, it is proved that multiple quadrotors can track the dynamic leader in a dynamic formation using the proposed protocol under formation feasibility and the existence of spanning trees in communication topology. Finally, a lightweight communication protocol is designed in the experimental platform. Several groups of visual simulations are carried out based on the Gazebo simulator. The proposed method is compared with two potential field methods in two scenarios through quantitative studies of task efficiency and error analysis. Overall, the completion time is improved by 86.02% and 31.98%, and the minimum tracking error is improved by 57.24% and 46.20% in scenario 1. The completion time is improved by 86.88% and 30.34%, and the minimum tracking errors are improved by 82.20% and 42.62% in scenario 2. Outdoor experiments on formation tracking are performed to verify the effectiveness of the proposed methods. Furthermore, several outdoor experiments are conducted in the same scenarios to validate the proposed algorithm.

Simultaneous Obstacle Avoidance and Target Tracking of Multiple Wheeled Mobile Robots With Certified Safety.

Collision avoidance plays a major part in the control of the wheeled mobile robot (WMR). Most existing collision-avoidance methods mainly focus on a single WMR and environmental obstacles. There are few products that cast light on the collision-avoidance between multiple WMRs (MWMRs). In this article, the problem of simultaneous collision-avoidance and target tracking is investigated for MWMRs working in the shared environment from the perspective of optimization. The collision-avoidance strategy is formulated as an inequality constraint, which has proven to be collision free between the MWMRs. The designed MWMRs control scheme integrates path following, collision-avoidance, and WMR velocity compliance, in which the path following task is chosen as the secondary task, and collision-avoidance is the primary task so that safety can be guaranteed in advance. A Lagrangian-based dynamic controller is constructed for the dominating behavior of the MWMRs. Combining theoretical analyses and experiments, the feasibility of the designed control scheme for the MWMRs is substantiated. Experimental results show that if obstacles do not threaten the safety of the WMR, the top priority in the control task is the target track task. All robots move along the desired trajectory. Once the collision criterion is satisfied, the collision-avoidance mechanism is activated and prominent in the controller. Under the proposed scheme, all robots achieve the target tracking on the premise of being collision free.

PACNav: a collective navigation approach for UAV swarms deprived of communication and external localization

This article proposes Persistence Administered Collective Navigation (PACNav) as an approach for achieving the decentralized collective navigation of unmanned aerial vehicle (UAV) swarms. The technique is based on the flocking and collective navigation behavior observed in natural swarms, such as cattle herds, bird flocks, and even large groups of humans. As global and concurrent information of all swarm members is not available in natural swarms, these systems use local observations to achieve the desired behavior. Similarly, PACNav relies only on local observations of the relative positions of UAVs, making it suitable for large swarms deprived of communication capabilities and external localization systems. We introduce the novel concepts of path persistence and path similarity that allow each swarm member to analyze the motion of other members in order to determine its own future motion. PACNav is based on two main principles: (a) UAVs with little variation in motion direction have high path persistence, and are considered by other UAVs to be reliable leaders; (b) groups of UAVs that move in a similar direction have high path similarity, and such groups are assumed to contain a reliable leader. The proposed approach also embeds a reactive collision avoidance mechanism to avoid collisions with swarm members and environmental obstacles. This collision avoidance ensures safety while reducing deviations from the assigned path. Along with several simulated experiments, we present a real-world experiment in a natural forest, showcasing the validity and effectiveness of the proposed collective navigation approach in challenging environments. The source code is released as open-source, making it possible to replicate the obtained results and facilitate the continuation of research by the community.

Coordinating transcription and replication to mitigate their conflicts in early Drosophila embryos.

Collisions between transcribing RNA polymerases and DNA replication forks are disruptive. The threat of collisions is particularly acute during the rapid early embryonic cell cycles of Drosophila when S phase occupies the entirety of interphase. We hypothesize that collision-avoidance mechanisms safeguard this early transcription. Real-time imaging of endogenously tagged RNA polymerase II (RNAPII) and a reporter for nascent transcripts in unperturbed embryos shows clustering of RNAPII at around 2min after mitotic exit, followed by progressive dispersal as associated nascent transcripts accumulate later in interphase. Abrupt inhibition of various steps in DNA replication, including origin licensing, origin firing, and polymerization, suppresses post-mitotic RNAPII clustering and transcription in nuclear cycles. We propose that replication dependency defers the onset of transcription so that RNAPII transcribes behind advancing replication forks. The resulting orderly progression can explain how early embryos circumvent transcription-replication conflicts to express essential developmental genes.

Multi-UAV flocking control with individual properties inspired by bird behavior

This paper focuses on the cooperative control problem for multiple unmanned aerial vehicles (UAVs). Unlike the existing models emphasizing the behavioral dimension, a general control framework inspired by bird behavior is developed by decoupling the problem into top decision-making and low-level management. Based on the individual properties, the models of communication modes, individual contribution, social relationship, and movement characteristics are proposed, where the individual contribution and social relationship can reflect the results of high-level decisions. The desired commands of the low-level realization are generated after the pre-linearized process and decoupled operator. Meanwhile, to guarantee the safety of each UAV, the collision avoidance mechanism is also applied. The convergence and stability of the proposed algorithm are also discussed. Experimental results demonstrate the effectiveness of the control framework in solving the UAV flocking control problem.

Constrained multi-agent ergodic area surveying control based on finite element approximation of the potential field

Heat Equation Driven Area Coverage (HEDAC) is a state-of-the-art multi-agent ergodic motion control guided by a gradient of a potential field. A finite element method is hereby implemented to obtain a solution of the Helmholtz partial differential equation, which models the potential field for surveying motion control. This allows us to survey arbitrarily shaped domains and to include obstacles in an elegant and robust manner intrinsic to HEDAC’s fundamental idea. For a simple kinematic motion, the obstacles and boundary avoidance constraints are successfully handled by directing the agent motion with the gradient of the potential. However, including additional constraints, such as the minimal clearance distance from stationary and moving obstacles and the minimal path curvature radius, requires further alternations of the control algorithm. We introduce a relatively simple yet robust approach for handling these constraints by formulating a straightforward optimization problem based on collision-free escape route maneuvers. This approach provides a guaranteed collision avoidance mechanism while being computationally inexpensive as a result of the optimization problem partitioning. The proposed motion control is evaluated in three realistic surveying scenarios simulations, showing the effectiveness of the surveying and the robustness of the control algorithm. Furthermore, potential maneuvering difficulties due to improperly defined surveying scenarios are highlighted and we provide guidelines on how to overpass them. The results are promising and indicate real-world applicability of the proposed constrained multi-agent motion control for autonomous surveying and potentially other HEDAC utilizations.

Clustering phenomenon of the singular Cucker–Smale model with finite communication weight and variable coupling strength

The clustering phenomenon of the Cucker–Smale type model has received lots of attention from many researchers. However, due to the difficulty of the study, to the best of our knowledge, there have been a few discussions about clustering only in the initial configuration or one dimension that induces very simple movements of agents.In this study, we deal with the initial configuration in which the agents of the Cucker–Smale type system in multidimensional form bi-clustering while exhibiting more complex movements without collision. For this, we design a singular interacting system of Cucker–Smale type with finite communication range and variable coupling strength, and provide a mathematical proof that the system leads to a situation in which two disjoint groups of agents pass without any collision between the agents and without changing the composition of the agents in each group. To show the proof, we first briefly discuss the system’s collision-avoidance mechanism and present sufficient conditions along with an interaction topology for an initial configuration of agents and an estimate of variable coupling strength leading to the emergence of flocking. Finally, we propose sufficient conditions to induce the desired phenomenon. We also provide the results of several numerical simulations performed to establish the significance of the conditions in our analytical results.

Contrasting coloured ventral wings are a visual collision avoidance signal in birds.

Collisions between fast-moving objects often cause severe damage, but collision avoidance mechanisms of fast-moving animals remain understudied. Particularly, birds can fly fast and often in large groups, raising the question of how individuals avoid in-flight collisions that are potentially lethal. We tested the collision-avoidance hypothesis, which proposes that conspicuously contrasting ventral wings are visual signals that help birds to avoid collisions. We scored the ventral wing contrasts for a global dataset of 1780 bird species. Phylogenetic comparative analyses showed that larger species had more contrasting ventral wings than smaller species, and that in larger species, colonial breeders had more contrasting ventral wings than non-colonial breeders. Evidently, larger species have lower manoeuvrability than smaller species, and colonial-breeding species frequently encounter con- and heterospecifics, increasing their risk of in-flight collisions. Thus, more contrasting ventral wing patterns in these species are a sensory mechanism that facilitates collision avoidance.