Collision Avoidance Scheme Research Articles

Soft formation control for unmanned surface vehicles under environmental disturbance using multi-task reinforcement learning

This paper proposes a distributed soft formation collision avoidance strategy to address the problem of formation obstacle avoidance in complex scenario under the environment interference where unmanned surface vehicles (USVs) are restricted in observation and possess no global reference frame for localisation. A multi-task training framework for formation control is also developed based on the motion characteristics of USVs and the leader–follower method. The soft actor–critic (SAC) reinforcement learning algorithm is adapted to construct agents. With elaborated auxiliary tasks, the soft formation algorithm demonstrates strong resilience to disturbances caused by unknown environmental loads or obstacle avoidance needs with only partial information about the environmental state, thus maintaining the formation shape. The proposed trajectory communication system and policy sharing mechanism allow USVs to approach the destination and avoid collisions whilst maintaining a noncompact formation that can be broken up temporarily to improve obstacle avoidance and can be restored quickly when the obstacle is avoided. Trained agents can be applied to different formations of varying sizes. Simulation results demonstrate the feasibility and effectiveness of the proposed method in different complex sea scenarios and its robustness to unknown environmental disturbances.

Optimization-Based Traffic Safety Improvement Strategy for Autonomous Vehicle Driving at Level Crossings

Railway level crossings pose a serious threat to the safety and mobility of drivers traversing them. Globally, a significant number of traffic violations occur annually at level crossings. This study proposes an optimization-based autonomous vehicle (AV) driving strategy to improve the traffic safety of AVs at level crossings. First, an optimization model considering the operational constraints is developed. The cost function of the proposed optimization-based driving strategy is the time difference between the time required by the train to reach the level crossing and the time taken by the AV to complete the whole crossing maneuver. Second, a MATLAB simulation model is developed to validate the proposed driving strategy. Simulation results show that the proposed driving strategy delivers optimized AV driving without stoppage. The comparative study also shows that the proposed method improves driving performance. For instance, when a train is 1 m away from the light signal while the AV’s initial speed is 36 km/h, the time required for the AV to reach the speed limit sign is about 0.44 s, while the time required by the train to reach the car block area is 0.13 s. Thus, the AV is forced to stop and then go after the train leaves the level crossing. Contrary to this, the proposed method allows a safe passage of the vehicle within 1.66 s without stoppage. Most importantly, the driving strategy is a collision avoidance strategy without unnecessary enforcement of stop-and-go when AVs have the chance of crossing safely. Consequently, the optimization-based driving strategy both improves traffic safety and reduces traveling time.

Walking characteristics and collision avoidance strategy in bidirectional pedestrian flow: a study focused on the influence of social groups

The walking characteristics and collision avoidance strategies in bidirectional pedestrian flow with both individuals and groups under three different pedestrian densities were investigated experimentally in the present study. The results showed that pedestrians formed lanes in bidirectional flow, and the lane number basically increased with pedestrian density. In addition, the lane number was positively related to the number of groups in the first row of the two opposite crowds. Compared with the no-group experiments, the walking speed of pedestrians showed less difference in the experiments with groups. The fundamental diagram of the bidirectional pedestrian flow with groups under different pedestrian density conditions was calculated. Three types of collision avoidance strategies, detouring around conflicting pedestrians, passing through the conflicting group in the middle, and group splitting, were summarised based on the experimental results. The statistical results of the three collision avoidance strategies adopted by different size groups under the low, medium, and high pedestrian density conditions were obtained. It was found that the large group was more likely to split into subgroups when they encountered conflicting pedestrians, while the small group was more likely to detour around the conflicting pedestrians. For the same group size, the larger the pedestrian density, the higher the proportion of groups that would choose to split into subgroups during the collision avoidance process.

Experimental and modelling studies of collision avoidance strategy choices and behavioural characteristics in interweaving pedestrian flow.

The mechanisms of collision avoidance (CA) behaviours in interweaving pedestrian flow movements are important for pedestrian space planning and emergency management but not well understood yet. In this paper, a series of controlled interweaving pedestrian flow experiments with different densities are carried out to investigate the CA behaviours, especially CA strategy choices. Four types of CA strategies are manually identified in these experiments. Nine characteristic parameters based on the trajectory data are defined to explore the characteristics of CA behaviours. The experimental results reveal that (i) the CA behaviours change with density levels; (ii) heterogeneities can be found for individual pedestrians; (iii) the defined characteristic parameters show different statistical features for different types of CA strategies, and correlations exist between most of the parameter pairs; (iv) it usually takes 0.5–2.5 s to complete a CA process with a trajectory length of 0.5–3.5 m. A multi-nomial logit (MNL) model and a long-short-term-memory (LSTM) model are established respectively for predicting pedestrians' choices of CA strategies using the selected characteristic parameters as inputs. The modelling results prove the importance of using time-series data for pedestrian behaviour modelling, and the LSTM models show advantages over the MNL model at this point.

Safe cooperative path following with relative-angle-based collision avoidance for multiple underactuated autonomous surface vehicles

This paper addresses the safe cooperative path following for multiple autonomous surface vehicles (ASVs) in the presence of static and dynamic obstacles. The positions and velocities of the obstacles are unknown, and only the relative angle is measured. The ASVs are subject to uncertain kinetics and unknown disturbances induced by wind, waves and currents. Switching cooperative guidance laws with collision avoidance are designed at the kinematic level, and a collision avoidance strategy based on the relative angle is introduced into the desired yaw rate design. A smooth switch between the cooperative path following and the collision avoidance is achieved by using an exponential-function-based weighted coefficient. At the kinetic level, nonlinear switching extended state observers (NSESOs) based on fractional power functions are developed to estimate the model uncertainties and environmental disturbances. Anti-disturbance kinetic control laws are designed based on the nonlinear switching extended state observers and finite-time convergent differentiators. A salient feature of the proposed cooperative path following method is that both the static and dynamic obstacles can be avoided by using the relative angle only. Through theoretical analysis, it is proven that the closed-loop system is input-to-state stable, and a safe distance can be maintained between the ASVs and the obstacles. Simulation and hardware-in-loop (HIL) experiment results are given to substantiate the effectiveness of the proposed safe cooperative path following control method for multiple ASVs based on relative angles.

Segmented trajectory planning strategy for active collision avoidance system

This paper presents a segmented trajectory planning strategy for active collision avoidance system. Considering the longitudinal and lateral movement of the obstacle vehicle, as well as the ego vehicle and obstacle outer contour limitations, the collision avoidance trajectory is divided into three segments: lane changing, overtaking and back to original lane. The starting point and end point of lane-change are decided based on longitudinal and lateral safety distance model according to the relative speed and distance as well as the outer contour of the two vehicles. Based on system objective function and lane-change trajectory cluster, vehicle states, dynamic constraints and vehicle body kinematics constraints, the optimal trajectory can be selected, which can monitor the relative location of the obstacle vehicle constantly and then ensure the vehicle can accomplish the collision avoidance safely and smoothly. Simulation and experiment results demonstrate the effectiveness and feasibility of proposed trajectory planning strategy for the active collision avoidance.

Collision Avoidance Strategy for Multivehicle Conflict on Common Rail

With the rapid expansion and application of bright logistics technology in the field of metallurgical production, in recent years, a type of intermediate granary storage and transportation system, signified by multiple child-parent rail car system, has gradually looked in steel plants, which effectively solves the drawbacks of crane craning in the intermediate warehouse area and improves the transportation competence of shop-level logistics. However, there are not many studies on conflict avoidance strategies for such multivehicle cooperative common rail transportation system in order to improve the ability of conflict detection and active avoidance in operation of such multivehicle transportation system. This paper draws on and improves the banker antideadlock conflict detection algorithm widely used in computer process management to detect conflicts in multivehicle on common rail system and effectively evades collision conflicts detected in the system by combining dynamic priority transfer with multiobjective time window offset. Finally, the validity of the improved banker algorithm conflict detection and priority transfer time window adjustment collision avoidance strategy is confirmed through an example analysis and comparison, providing a security guarantee for the efficient and intelligent operation of such common rail multivehicle logistics system, particularly for the exploration of the logistics common rail multivehicle collision avoidance strategy in metallurgical heavy-duty workshop.

Integrated-Hybrid Framework for Connected and Autonomous Vehicles Microscopic Traffic Flow Modelling

In this study, a novel traffic flow modeling framework is proposed considering the impact of driving system and vehicle mechanical behavior as two different units on the traffic flow. To precisely model the behavior of Connected and Autonomous (CA) vehicles, three submodels are proposed as a novel microscopic traffic flow framework, named Integrated-Hybrid (IH) model. Focusing on the realization of the car following behavior of CA vehicles, the driving system (vehicle control system) and the vehicle mechanical system are modeled separately and linked by throttle and brake actuators model. The IH model constitutes the key part of the Full Velocity Difference (FVD) model considering the mechanical capability of vehicles and dynamic collision avoidance strategies to ensure the safety of following distance between two consecutive vehicles. Linear stability conditions are derived for each model and developing methodology for each submodel is discussed. Our simulations revealed that the IH model successfully generates velocity and acceleration profiles during car following maneuvers and throttle angle/brake information in connected vehicles environment can effectively improve traffic flow stability. The vehicles’ departure and arrival process while passing through a signal-lane with a traffic light considering the anticipation driving behavior and throttle angle/brake information of direct leading vehicle was explored. Our numerical results demonstrated that the IH model can capture the velocity fluctuations, delay times, and kinematic waves efficiently in traffic flow.

Polygon formation of multiple nonholonomic mobile robots with double-level-control collision avoidance scheme

This paper considers a polygon formation control of multiple robots with nonholonomic constraints enclosing a goal target and double-level-control collision avoidance scheme. Double-level-control scheme consisted of upper-level and lower-level controls are proposed for trajectory generation and tracking control of multi-robot systems. Both upper-level and lower-level controls operate collision avoidance mechanisms based on potential functions. The proposed control scheme guarantees that the group of robots are kept in the polygon formation and driven to a goal, while avoiding collisions during the travel. Moreover, the designed interaction between the upper- and lower-level controls guarantees that the mobile robots are not trapped in local minima or deadlock case. Experiments of the formation of three-robots are conducted to show the performance of the mobile robots in accomplishing a polygon formation while achieving the goal without any collision and no local minima.

Self-Organizing Cooperative Pursuit Strategy for Multi-USV with Dynamic Obstacle Ships

A self-organizing cooperation strategy for multiple unmanned surface vessels (USVs) to pursue intelligent evaders in the case of a dynamic obstacle vessel is proposed. Firstly, a self-organizing cooperative hunting strategy is proposed to form an Apollonius circle. According to the escape strategies of evaders under different encirclement states, the pursuers are divided into pursuit group and ambush group. The pursuit group drives the evaders into the ambush area and completes the encirclement together with the ambush group. In order to better deal with the dynamic obstacle ships encountered in the pursuit process in the dynamic ocean environment, the artificial potential field-based collision avoidance method for inter-USV and dynamic collision avoidance strategy for encountering obstacle ships based on International Regulations for Preventing Collisions at Sea (COLREGs) are proposed. The simulation results show that the algorithm can make the pursuers complete the encirclement of the evaders and has good obstacle avoidance performance and flexibility in the environment with dynamic obstacle ships.

A Systematic Review of Autonomous Emergency Braking System: Impact Factor, Technology, and Performance Evaluation

In order to track the research progress of AEB-related technologies, this paper makes a systematic analysis and research on the impact factors, key technologies, and effect evaluation of AEB. First, the paper deeply analyzes the three levels of factors affecting the performance of AEB, which are vehicle factors, driver factors, and environmental factors. Second, the paper deeply studies the technical status of the three subsystems of environment perception, decision-making, and control execution. Particularly, the performance of Mazda, Honda, NHTSA, Berkeley, and Seungwuk Moon are compared and analyzed based on MATLAB. Third, the paper summarizes the current AEB virtual test methods, closed field test methods, and its test sites. Three classic evaluation methods in the world, including the AEB test evaluation standards of ENCAP, IIHS, and i-Vista are analyzed. Finally, the paper prospects the specific research directions, including the protection of vulnerable road users, target detection method, collision avoidance strategy, complex scenarios application, and application of emerging technologies.

Codesign of dynamic collision avoidance and trajectory tracking for autonomous surface vessels with nonlinear model predictive control

In this paper, a predictive control scheme for dynamic collision avoidance and formation trajectory tracking of autonomous surface vessels (ASVs) was designed. First, a trajectory-tracking nonlinear controller for ASVs was developed in the framework of model predictive control (MPC). Next, to realize leader-follower formation control, a graph topology approach was proposed to maintain a balanced distance between the leader and followers. Then, a set of nonlinear dynamic collision avoidance constraints based on the time to closest point of approach (TCPA) and distance to closest point of approach (DCPA) models was applied as control constraints. Specifically, the ship collision risk index (CRI) was proposed as a collision avoidance constraint for the controller. Simulation studies with various collision avoidance scenarios and predictive horizons were conducted using the dynamic CyberShip II model. The simulation results demonstrate the effectiveness of the proposed codesign scheme for formation tracking control and dynamic collision avoidance. Additionally, the simulation results show that the calculation time of the system was greatly reduced when using a trigger function, with an average calculation time of approximately 0.251 s.

Bearing‐based autonomous communication relay positioning under field‐of‐view constraints

AbstractThis article investigates the problem of communication relay establishment for multiple agent‐based mobile units using a relay vehicle. The objective is to drive autonomously the relay vehicle to attain a position for communication relay establishment while maintaining the other vehicles inside of its field‐of‐view. A bearing‐based control law is proposed for the relay drone and designed for both single and multiple agents. We also provide a collision avoidance scheme that ensures no collisions between the relay and other agents. Numerical simulations and experimental results are reported as well to show the efficacy of the proposed approach.

Autonomous obstacle avoidance strategies in the mission of large space debris removal using potential function

Debris removal problem has become a consensus of international communities with the dramatically increase of space debris. When the service spacecraft (Servicer) carries out debris removal operations, there may exist collision threaten from other small objects which fly around the target debris. In this paper, autonomous collision avoidance strategies using improved potential function are studied, which drive the Servicer to rendezvous with the large debris and avoid obstacles around the trajectory. Firstly, the repulsive potential for static obstacles is constructed based on the relative position vector. Considering the necessary of avoidance of dynamic obstacles, the relative velocity vector between the Servicer and dynamic obstacles in space is introduced to design the repulsive potential for dynamic obstacles. Also, the corresponding control law is derived to drive the Servicer to approach the large debris. Finally, numerical simulations are performed to validate the effectiveness and stability of the obstacle avoidance strategy for debris removal.

Formation Tracking and Obstacle Avoidance for Multiple Quadrotors With Static and Dynamic Obstacles

This letter proposes a novel distributed cooperative control algorithm to address the problem of collision avoidance and obstacle avoidance for multiple quadrotors during the formation tracking process. The proposed algorithm couples collision avoidance and obstacle avoidance schemes into the control layer. To avoid collisions between quadrotors in time, a repulsion function based on Hooke’s law with damping is proposed, which fully considers the relative position and relative velocity between quadrotors. In addition, based on the obstacle avoidance behavior of pigeons, a split-merge strategy is designed for multiple quadrotors to avoid static and dynamic obstacles. The split-merge strategy is driven by the relative position between the quadrotors and the obstacles, and it can calculate the optimal velocity to keep the quadrotors away from obstacles in the field of view. Several simulations and outdoor experiments for multiple quadrotors are presented to verify the effectiveness of the theoretical results.

Collision Avoidance Strategy for Self-Driving Vehicles Based on Drivers and Environmental Factors

In order to enhance the driving safety, various driving assistant systems are gradually developed and widely used. Based on the safe distance model related to the braking process, one warning/dangerous critical distance model for longitudinal control is proposed in this paper, considering the driver’s style and environment factors. The variable driver response time is obtained by employing fuzzy control theory. The author then establishes the vehicle dynamics model and its inverse model. Through the relative distance error and relative speed error, the upper controller is designed to obtain the expected acceleration based on sliding mode control, and the lower controller based on PID control is designed to track the acceleration. In order to solve the chattering problem of sliding mode controller, the author proposes variable gain coefficient control method. The fixed gain parameter in the expected acceleration formula is related to the negative exponential correlation formula, which can effectively depress the chattering of the expected acceleration. Finally, the author uses CarSim and Simulink to establish a co-simulation model of the collision avoidance warning system, and designs the test simulation conditions for typical traffic test scenarios to verify the control algorithm effectiveness.

Decentralized Multi-Robot Collision Avoidance: A Systematic Review from 2015 to 2021

An exploration task can be performed by a team of mobile robots more efficiently than human counterparts. They can access and give live updates for hard-to-reach areas such as a disaster site or a sewer. However, they face some issues hindering them from optimal path planning due to the symmetrical shape of the environments. Multiple robots are expected to explore more areas in less time while solving robot localization and collision-avoidance issues. When deploying a multi-robot system, it is ensured that the hardware parts do not collide with each other or the surroundings, especially in symmetric environments. Two types of methods are used for collision avoidance: centralized and decentralized. The decentralized approach has mainly been used in recent times, as it is computationally less expensive. This article aims to conduct a systematic literature review of different collision-avoidance strategies and analyze the performance of innovative collision-avoidance techniques. Different methods such as Reinforcement Learning (RL), Model Predictive Control (MPC), Altruistic Coordination, and other approaches followed by selected studies are also discussed. A total of 17 studies are included in this review, extracted from seven databases. Two experimental designs are studied: empty/open space and confined indoor space. Our analysis observed that most of the studies focused on empty/open space scenarios and verified the proposed model only through simulation. ORCA is the primary method, against which all the state-of-the-art techniques are evaluated. This article provides a comparison between different methods used for multi-robot collision avoidance. It discusses if the methods used are focused on safety or path planning. It also sheds light on the limitations of the studies included and possible future directions.

Quasi-critical collision-avoidance strategy for autonomous vehicles in complex traffic scenarios based on exclusive area of relative velocity vector algorithm

Collision avoidance is one of the most important requirements for autonomous vehicles, particularly in complex and congested traffic scenarios where trajectories have little safety redundancy. However, simultaneously reaching the required accuracy and universal feasibility for different collision-avoidance behaviours is difficult due to the multi-state coupled motion of vehicles. To achieve the maximum traversability and ensure the safety of autonomous vehicles in any complex scenarios, we propose a quasi-critical collision-avoidance strategy based on a newly developed algorithm: the exclusive area-of-relative-velocity vector. This strategy first involves the construction of an exclusive area-of-velocity vector for each object vehicle to extract its position relative to the subject vehicle. In this procedure, to establish a subject-motion-decoupled scenario, projective transformation is applied to regularise the moving elliptical contour of the subject vehicle as a settled circle while retaining all positional relationships between the subject and object vehicles using the invariants. Subsequently, a group of escaping conditions for this exclusive area are established to express this quasi-critical collision-avoidance strategy explicitly and mathematically. The ultimate ability to escape from such an area is determined through theoretical derivations and experiments according to vehicle dynamics. In terms of real scenario data, a set of escaping equations is established to calculate the escaping conditions subject to the current state and the ultimate motion ability. Via scenario verifications, this strategy is shown to represent the safety boundary accurately and ensure quasi-critical collision-avoidance conditions under complex scenarios.

Bio-Inspired Collision Avoidance in Swarm Systems via Deep Reinforcement Learning

Autonomous vehicles have been highlighted as a major growth area for future transportation systems and the deployment of large numbers of these vehicles is expected when safety and legal challenges are overcome. To meet the necessary safety standards, effective collision avoidance technologies are required to ensure that the number of accidents are kept to a minimum. As large numbers of autonomous vehicles, operating together on roads, can be regarded as a swarm system, we propose a bio-inspired collision avoidance strategy using virtual pheromones; an approach that has evolved effectively in nature over many millions of years. Previous research using virtual pheromones showed the potential of pheromone-based systems to maneuver a swarm of robots. However, designing an individual controller to maximise the performance of the entire swarm is a major challenge. In this paper, we propose a novel deep reinforcement learning (DRL) based approach that is able to train a controller that introduces collision avoidance behaviour. To accelerate training, we propose a novel sampling strategy called <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Highlight Experience Replay</i> and integrate it with a Deep Deterministic Policy Gradient algorithm with noise added to the weights and biases of the artificial neural network to improve exploration. To evaluate the performance of the proposed DRL-based controller, we applied it to navigation and collision avoidance tasks in three different traffic scenarios. The experimental results showed that the proposed DRL-based controller outperformed the manually-tuned controller in terms of stability, effectiveness, robustness and ease of tuning process. Furthermore, the proposed Highlight Experience Replay method outperformed than the popular Prioritized Experience Replay sampling strategy by taking 27% of training time average over three stages.

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

In order to realize the autonomous collision avoidance of unmanned surface vehicles (USVs), an intelligent hybrid collision avoidance algorithm based on deep reinforcement learning is proposed. First, the navigation situation model of the USV is designed, the geometric model of the encounter between two ships is established based on navigation practice. According to static and dynamic obstacles, hybrid risk assessment and collision avoidance model are proposed, the risk factor is calculated. Then, for static obstacles, the collision cone is introduced, for dynamic ships, the COLREGS is observed, the encounter situation is quantified into five types. Collision avoidance strategy is formulated. Finally, the state, action, reward function and network structure are designed. Aiming at the problem of low utilization for samples in random sampling, this paper improves the original sampling mechanism of DDPG, and the priority sampling mechanism with cumulative pruning is proposed in this paper. Simulation experiments are carried out in several typical encounter scenarios. The results show that this algorithm can accurately judge the encounter situation, give reasonable collision avoidance actions, and realize effective collision avoidance in a complex environment with dynamic and static obstacles. The research can provide theoretical basis and method reference for autonomous navigation of USVs.