Optimal Collision Research Articles

Formation Control With Collision Avoidance Through Deep Reinforcement Learning Using Model-Guided Demonstration.

Generating collision-free, time-efficient paths in an uncertain dynamic environment poses huge challenges for the formation control with collision avoidance (FCCA) problem in a leader-follower structure. In particular, the followers have to take both formation maintenance and collision avoidance into account simultaneously. Unfortunately, most of the existing works are simple combinations of methods dealing with the two problems separately. In this article, a new method based on deep reinforcement learning (RL) is proposed to solve the problem of FCCA. Especially, the learning-based policy is extended to the field of formation control, which involves a two-stage training framework: an imitation learning (IL) and later an RL. In the IL stage, a model-guided method consisting of a consensus theory-based formation controller and an optimal reciprocal collision avoidance strategy is designed to speed up training and increase efficiency. In the RL stage, a compound reward function is presented to guide the training. In addition, we design a formation-oriented network structure to perceive the environment. Long short-term memory is adopted to enable the network structure to perceive the information of obstacles of an uncertain number, and a transfer training approach is adopted to improve the generalization of the network in different scenarios. Numerous representative simulations are conducted, and our method is further deployed to an experimental platform based on a multiomnidirectional-wheeled car system. The effectiveness and practicability of our proposed method are validated through both the simulation and experiment results.

Real-time collision avoidance planning for unmanned surface vessels based on field theory

With the development of commercial application of unmanned surface vessels, the real-time collision avoidance is the research focus on the multi-vessels situation, the structure complexity and time consumption problem limit its efficiency. To simplify the complex model and enhance the collision avoidance performance, the field theory is used to describe the navigation of ships abstractly, from which a more simplified optimization model of collision avoidance is derived, so the synthetic field regarding the virtual spatial electric field and velocity field is presented to exhibit the situation of ship motion. On the basis of the field environment, the more simplified collision avoidance model is designed, combining with the improved Electromagnetism-like mechanism algorithm, to generate the optimal collision avoidance strategy, for considering the energy and other loss. The complete algorithm has been compared and validated under the different layout, it is worth mentioning that the algorithm can potentially be developed to advance collision-free navigation in multi-vessels situation.

GLAS: Global-to-Local Safe Autonomy Synthesis for Multi-Robot Motion Planning With End-to-End Learning

We present GLAS: Global-to-Local Autonomy Synthesis, a provably-safe, automated distributed policy generation for multi-robot motion planning. Our approach combines the advantage of centralized planning of avoiding local minima with the advantage of decentralized controllers of scalability and distributed computation. In particular, our synthesized policies only require relative state information of nearby neighbors and obstacles, and compute a provably-safe action. Our approach has three major components: i) we generate demonstration trajectories using a global planner and extract local observations from them, ii) we use deep imitation learning to learn a decentralized policy that can run efficiently online, and iii) we introduce a novel differentiable safety module to ensure collision-free operation, thereby allowing for end-to-end policy training. Our numerical experiments demonstrate that our policies have a 20% higher success rate than optimal reciprocal collision avoidance, ORCA, across a wide range of robot and obstacle densities. We demonstrate our method on an aerial swarm, executing the policy on low-end microcontrollers in real-time.

Cooperative Optimal Collision Avoidance Laws for a Hybrid-Tailed Robotic Fish

This brief addresses the problem of collision avoidance by robotic fish that have a single caudal fin. The fish is hybrid-tailed in that the caudal fin is driven by a double-jointed mechanism. One joint is driven by a servomotor and the second joint is driven by an ionic polymer-metal composite (IPMC) actuator, which is often called an artificial muscle. For this type of robotic fish, collision avoidance is performed by employing a collision cone approach. Within the framework of the collision cone approach, Lyapunov-based methods are employed to determine analytical expressions of nonlinear guidance laws with which cooperative collision avoidance can be achieved. It is shown how these cooperative collision avoidance laws can be made optimal in the sense of minimizing an energy-like performance index. The conditions under which the developed guidance laws are robust to sensor measurement errors are determined. Simulations and experiments are performed to validate the guidance laws.

UAV environmental perception and autonomous obstacle avoidance: A deep learning and depth camera combined solution

In agriculture, Unmanned Aerial Vehicles (UAVs) have shown great potential for plant protection. Uncertain obstacles randomly distributed in the unstructured farmland usually pose significant collision risks to flight safety. In order to improve the UAV’s intelligence and minimize the obstacle’s adverse impacts on operating safety and efficiency, we put forward a comprehensive solution which consists of deep-learning based object detection, image processing, RGB-D information fusion and Task Control System (TCS). Taking full advantages of both deep learning and depth camera, this solution allows the UAV to perceive not only the presence of obstacles, but also their attributes like category, profile and 3D spatial position. Based on the object detection results, collision avoidance strategy generation method and the corresponding calculation approach of optimal collision avoidance flight path are elaborated detailly. A series of experiments are conducted to verify the UAV’s environmental perception ability and autonomous obstacle avoidance performance. Results show that the average detection accuracy of CNN model is 75.4% and the mean time cost for processing single image is 53.33 ms. Additionally, we find that the prediction accuracy of obstacle’s profile and position depends heavily on the relative distance between the object and the depth camera. When the distance is between 4.5 m and 8.0 m, errors of object’s depth data, width and height are −0.53 m, −0.26 m and −0.24 m respectively. Outcomes of simulation flight experiments indicated that the UAV can autonomously determine optimal obstacle avoidance strategy and generate distance-minimized flight path based on the results of RGB-D information fusion. The proposed solution has extensive potential to enhance the UAV’s environmental perception and autonomous obstacle avoidance abilities.

A Novel Collision-Free Navigation Approach for Multiple Nonholonomic Robots Based on ORCA and Linear MPC

In the study of collision-free navigation methods of multirobots, much attention has been paid to the constraints of external environment. However, most of the wheeled mobile robots are subjected to nonholonomic constraints. A collision between robots may occur if the nonholonomic constraints are neglected. This paper presents an improved approach to collision-free navigation for multi-nonholonomic robots. This approach combines the Optimal Reciprocal Collision Avoidance (ORCA) algorithm and Model Predictive Control (MPC) strategy. ORCA used a simple robot model, in which kinematics and dynamics are ignored. To cope with this problem, the MPC controller is introduced. In each ORCA step, the reference trajectory, reference control inputs, and “safe zones” are generated based on the new velocity. Consequently, the derived safe zone is transformed into the constraints of decision variables for a MPC controller. Finally, quadratic programming is used to solve the MPC problem by successive linearization of an error model of the mobile robot. Simulation results illustrate the effectiveness of the proposed method.

Realistic Multi-Agent Formation Using Discretionary Group Behavior (DGB)

Simulating groups and their behaviors have been one of the important topics recently. This paper proposes a novel velocity-based method to simulate the realistic behavior of groups moving in a specific formation in a virtual environment including other groups and obstacles. The proposed algorithm, we called “DGB—Discretionary Group Behavior”, takes advantage of ORCA (Optimal Reciprocal Collision Avoidance) half-planes for both grouping and collision avoidance strategy. By considering new half-planes for each agent, we can have more reasonable and intelligent behavior in front of challenging obstacles and other agents. Unlike recent similar works, independent members in a group do not have predefined connections to each other even though they can keep the group’s formation while moving and trying to follow their best neighbors discretionarily in critical situations. Through experiments, we found that the proposed algorithm can yield more human-like group behavior in a crowd of agents.

Multi-robot path planning using improved particle swarm optimization algorithm through novel evolutionary operators

The highlight of this paper is to propose an innovative approach to compute an optimal collision free trajectory path for each robot in a known and complex environment. The problem under consideration has been solved by employing an improved version of particle swarm optimization (IPSO) with evolutionary operators (EOPs). In the present context, PSO is improved with the concept of governance in human society and two evolutionary operators such as multi-crossover inherited from the genetic algorithm, and bee colony operator to enhance the intensification capability of the IPSO algorithm. The algorithm proposed to compute the deadlock free subsequent coordinate of an individual robot from their present coordinate, in addition, to minimize the path length for each robot by maintaining a good balance between intensification and diversification. Results obtained from the proposed IPSO-EOPs have been compared with competitors such as DE and IPSO in a similar environment to substantiate the robustness and usefulness of the algorithm. It perceives from the result obtained from simulation and experimentation that IPSO-EOPs is succeeding IPSO, and DE in terms of arrival time, generating a safe optimal path, and energy utilization during the travel.

Local motion simulation using deep reinforcement learning

AbstractTraditional local motion simulation focuses largely on avoiding collisions in the next frame. However, due to its lack of forward looking, the global trajectory of agents usually seems unreasonable. As a method of optimizing the overall reward, deep reinforcement learning (DRL) can better correct the problems that exist in the traditional local motion simulation method. In this article, we propose a local motion simulation method integrating optimal reciprocal collision avoidance (ORCA) and DRL, referred to as ORCA‐DRL. The main idea of ORCA‐DRL is to perform local collision avoidance detection via ORCA and smooth the trajectory at the same time via DRL. We use a deep neural network (DNN) as the state‐to‐action mapping function, where the state information is detected by virtual visual sensors and the action space includes two continuous spaces: speed and direction. To improve data utilization and speed up the training process, we use the proximal policy optimization based on the actor–critic (AC) framework to update the DNN parameters. Three scenes (circle, hallway, and crossing) are designed to evaluate the performance of ORCA‐DRL. The results reveal that, compared with the ORCA, our proposed ORCA‐DRL method can: (a) reduce the total number of frames, leading to less time for agents to reach their destination; and (b) effectively avoid local optima, evidenced by smoothed global trajectories.

Visual Flight Rules-Based Collision Avoidance Systems for UAV Flying in Civil Aerospace

The operation of Unmanned Aerial Vehicles (UAVs) in civil airspace is restricted by the aviation authorities, which require full compliance with regulations that apply for manned aircraft. This paper proposes control algorithms for a collision avoidance system that can be used as an advisory system or a guidance system for UAVs that are flying in civil airspace under visual flight rules. A decision-making system for collision avoidance is developed based on the rules of the air. The proposed architecture of the decision-making system is engineered to be implementable in both manned aircraft and UAVs to perform different tasks ranging from collision detection to a safe avoidance manoeuvre initiation. Avoidance manoeuvres that are compliant with the rules of the air are proposed based on pilot suggestions for a subset of possible collision scenarios. The proposed avoidance manoeuvres are parameterized using a geometric approach. An optimal collision avoidance algorithm is developed for real-time local trajectory planning. Essentially, a finite-horizon optimal control problem is periodically solved in real-time hence updating the aircraft trajectory to avoid obstacles and track a predefined trajectory. The optimal control problem is formulated in output space, and parameterized by using B-splines. Then the optimal designed outputs are mapped into control inputs of the system by using the inverse dynamics of a fixed wing aircraft.

Nature inspired evolutionary approaches for robot navigation: Survey

Autonomous robot navigation (ARN) requires intelligent systems equipped with human capabilities to achieve controlled motion among flats, ridgesor obstacles while avoiding collision in known or partially unknown environment with safety of equipment and manpower involved. Its objective is to generate optimal collision free path from initial to target location within minimal time and energy requirement. There are various navigational approaches (classical, heuristics, hybrid or nature inspired) and its choice depends on type of environment (static or dynamic) and application area. Nature inspired evolutionary (NIE) approaches is an emerging paradigm to develop robust and dynamic problems to address non-linear or multi-objective problemsin dynamic environment. This work is an attempt to review recently developed NIE approaches based on different biological groups for application in field of robot path planning (RPP).

DCAD: Decentralized Collision Avoidance With Dynamics Constraints for Agile Quadrotor Swarms

We present DCAD, a novel, decentralized collision avoidance algorithm for navigating a swarm of quadrotors in dense environments populated with static and dynamic obstacles. Our algorithm relies on the concept of Optimal Reciprocal Collision Avoidance (ORCA) and utilizes a flatness-based Model Predictive Control (MPC) to generate local collision-free trajectories for each quadrotor. We feedforward linearize the non-linear dynamics of the quadrotor and subsequently use this linearized model in our MPC framework. Our approach tends to compute safe trajectories that avoid quadrotors from entering each other's downwash regions during close proximity maneuvers. In addition, we account for the uncertainty in the position and velocity sensor data using Kalman filter. We evaluate the performance of our algorithm with other state-of-the-art decentralized methods and demonstrate its superior performance in terms of smoothness of generated trajectories and lower probability of collision during high-velocity maneuvers.

Deterministic-Based Ship Anti-Collision Route Optimization with Web-Based Application

Most of the accidents are caused by human error at sea so, decision making process made by navigators should be more computerised and automated. The supported decision making can be a step forward to decrease the risk of collision. This paper, in this respect, aims to present a deterministic approach to support optimum collision avoidance trajectory. This approach involves a collision avoidance course alteration. A web-based application coded with "JavaScript" programming language on the "Processing" software platform which allows the own ship to change her course in a deterministic manner to avoid collision optimally has been introduced. Algorithm structure of the method has been formulated and organized according to the International Regulation for Preventing Collision at Sea (COLREGs). The experimental tests results have revealed that the system is practicable and feasible and considerably outperforms heuristic-based method. It is thought that the developed method can be applied in an intelligent avoidance system on board and provides contribution to ship collision avoidance process, automation of ship motion control and ship traffic engineering.

Retention Time and Optimal Collision Energy Advance Structural Annotation Relied on LC-MS/MS: An Application in Metabolite Identification of an Antidementia Agent Namely Echinacoside.

The structural annotation of metabolites now relies heavily on HR-MS/MS information, resulting in ambiguous identities in most cases. More auxiliary evidence is therefore desired to achieve confirmative identification. Herein, we made an attempt to involve retention time (tR) along with optimal collision energy (OCE) as the additionally structural clues, and the applicability validation was conducted via confidence-enhanced metabolite characterization of echinacoside, an antidementia drug candidate within clinical trials. Quantitative structure-retention relationships (QSRR) were modeled via assaying 184 authentic compounds on RPLC, HILIC, and serially coupled RPLC and HILIC (RPLC-HILIC). Online energy-resolved MS was developed to yield breakdown graphs for selected ion transitions, and OCE was demonstrated to be superior to CE50 toward pointedly denoting the bonds-of-interest. Nineteen metabolites (M1-M19) were confidently identified in biological samples from echinacoside-treated rats by analyzing m/z values first to yield empirical formulas and substructures, and tR and OCE subsequently contributed to sift the candidate structures. Structural identification was validated by oral administration of three relevant compounds in parallel and chromatographic purification as well. Above all, the integration of retention and dissociation behaviors enabled promoting one step forward for structural annotation confidences merely relied on HR-MS/MS.

LSwarm: Efficient Collision Avoidance for Large Swarms With Coverage Constraints in Complex Urban Scenes

In this letter, we address the problem of collision avoidance for a swarm of UAVs used for continuous surveillance of an urban environment. Our method, LSwarm, efficiently avoids collisions with static obstacles, dynamic obstacles and other agents in three-dimensional urban environments while considering coverage constraints. LSwarm calculates collision avoiding velocities that maximize the conformity of an agent to an optimal path given by a global coverage strategy and ensure sufficient resolution of the coverage data collected by each agent. Our algorithm is formulated based on optimal reciprocal collision avoidance and is scalable with respect to the size of the swarm. We evaluate the coverage performance of LSwarm in realistic simulations of a swarm of quadrotors in complex urban models. In practice, our approach can compute collision avoiding velocities for a swarm composed of tens to hundreds of agents in a few milliseconds on dense urban scenes consisting of tens of buildings.

Optimal Collision Energies and Bioinformatics Tools for Efficient Bottom-up Sequence Validation of Monoclonal Antibodies.

Rigorous validation of amino acid sequence is fundamental in the characterization of original and biosimilar protein biopharmaceuticals. Widely accepted workflows are based on bottom-up mass spectrometry, and they often require multiple techniques and significant manual work. Here, we demonstrate that optimization of a set of tandem mass spectroscopy (MS/MS) collision energies and automated combination of all available information in the measurements can increase the sequence validated by one technique close to the inherent limits. We created a software (called "Serac") that consumes results of the Mascot database search engine and identifies the amino acids validated by bottom-up MS/MS experiments using the most rigorous, industrially acceptable definition of sequence coverage (we term this "confirmed sequence coverage"). The software can combine spectra at the level of amino acids or fragment ions to exploit complementarity, provides full transparency to justify validation, and reduces manual effort. With its help, we investigated collision energy dependence of confirmed sequence coverage of individual peptides and full proteins on trypsin-digested monoclonal antibody samples (rituximab and trastuzumab). We found the energy dependence to be modest, but we demonstrated the benefit of using spectra taken at multiple energies. We describe a workflow based on 2-3 LC-MS/MS runs, carefully selected collision energies, and a fragment ion level combination, which yields ∼85% confirmed sequence coverage, 25%-30% above that from a basic proteomics protocol. Further increase can mainly be expected from alternative digestion enzymes or fragmentation techniques, which can be seamlessly integrated to the processing, thereby allowing effortless validation of full sequences.

Optimal collision energy is an eligible molecular descriptor to boost structural annotation: An application for chlorogenic acid derivatives-focused chemical profiling

Tandem mass spectrometry (MS/MS), in particular high-resolution MS/MS, is able to provide element compositions and substructures for the detected signals. However, it is still challenging to configure the whole structures via linking those substructures. Efforts were devoted here to propose and validate optimal collision energy (OCE) to be an auxiliary structural clue to mass-to-charge ratios (m/z), and online energy-resolved MS was developed to yield OCEs. Chlorogenic acid derivatives (CADs) were utilized as the proof-of-concept because diverse isomers usually initiated by the different linkage manners between the quinic acid/shikimic acid and cinnamoyl substituents(s), i.e. caffeoyl group, coumaroyl group, etc. Liquid chromatography–hybrid ion trap-time of flight MS (LC–IT-TOF–MS) was implemented to capture CADs in two well-known herbal medicines namely Lonicerae japonicae Flos and Inulae Flos. Afterwards, hybrid triple quadrupole-linear ion trap MS (Qtrap-MS) was deployed to acquire OCEs for the primary fragmentation pathways of all detected CADs through online energy-resolved MS. On the other side, structural calculations were conducted to figure out the relationships between OCEs and bond properties. Isomeric differences occurred for OCEs, and LC elution program as well as ionization parameters could not affect OCEs. Twenty-four and thirty-one CADs were hunted and putatively identified by LC–IT-TOF-MS in Lonicerae japonicae Flos and Inulae Flos, respectively, and the structural annotation was advanced by applying the OCE-bond property relationships. To verify the structures, CADs-of-interest were purified from Lonicerae japonicae Flos using an automated fraction collector and definitely identified with NMR spectroscopy. Exact consistence occurred for the structural identification of mono-caffeoylquinic acid isomers between LC–MS/MS and NMR analyses. Consequently, OCE is an inherent physicochemical parameter of a given compound and is an eligible structural descriptor to offset the ability of LC–MS/MS towards the chemical profiling of complex matrices.

A velocity obstacles approach for autonomous landing and teleoperated robots

Velocity obstacles (VO) are one of the most successful methods to compute collision-free trajectory for multi-agent systems. VO provide for each autonomous robot the set of velocities that avoids collisions with other robots (sharing or not the same motion policy) and with moving or static obstacles in the environment. In this paper we will focus on a particular efficient implementation of the VO paradigm available in the literature, called optimal reciprocal collision avoidance. After highlighting and solving a couple of deadlock situations that the current implementation cannot manage, we extend this approach to two challenging applications: (1) the landing of a UAV onto a UGV in crowded environments, and (2) the generation of force feedback for teleoperated vehicles. The theoretical outcomes are validated in simulated scenarios using V-REP as a virtual robot development tool.

Deciphering Isomers with a Multiple Reaction Monitoring Method for the Complete Detectable O-Glycan Repertoire of the Candidate Therapeutic, Lubricin.

Glycosylation is a fundamental post-translational modification, occurring on half of all proteins. Despite its significance, our understanding is limited, in part due to the inherent difficulty in studying these branched, multi-isomer structures. Accessible, detailed, and quantifiable methods for studying glycans, particularly O-glycans, are needed. Here we take a multiple reaction monitoring (MRM) approach to differentiate and relatively quantify all detectable glycans, including isomers, on the heavily O-glycosylated protein lubricin. Lubricin (proteoglycan 4) is essential for lubrication of the joint and eye. Given the therapeutic potential of lubricin, it is essential to understand its O-glycan repertoire in biological and recombinantly produced samples. O-Glycans were released by reductive β-elimination and defined, showing a range of 26 neutral, sulfated, sialylated, and both sulfated and sialylated core 1 (Galβ1-3GalNAcα1-) and core 2 (Galβ1-3(GlcNAcβ1-6)GalNAcα1-) structures. Isomer-specific MRM transitions allowed effective differentiation of neutral glycan isomers as well as sulfated isomeric structures, where the sulfate was retained on the fragment ions. This strategy was not as effective with labile sialylated structures; instead, it was observed that the optimal collision energy for the m/z 290.1 sialic acid B-fragment differed consistently between sialic acid isomers, allowing differentiation between isomers when fragmentation spectra were insufficient. This approach was also effective for purchased Neu5Acα2-3Galβ1-4Glc and Neu5Acα2-6Galβ1-4Glc and for Neu5Acα2-3Galβ1-4GlcNAc and Neu5Acα2-6Galβ1-4GlcNAc linkage isomers with the Neu5Acα2-6 consistently requiring more energy for optimal generation of the m/z 290.1 fragment. Overall, this method provides an effective and easily accessible approach for the quantification and annotation of complex released O-glycan samples.

Chemical profiling for bile acid derivatives in yak bile

Bile acids( BAs),the major constituents of bile,are also known to be potential biomarkers of various diseases,especially liver disease. The systematic analysis of BAs is believed to be of great importance towards the clarification of the effective material basis for bile-type medicines,and the diagnosis and therapy of related diseases as well. As a part of systematic study on bile-type medicine ongoing in our group,this study lays emphasis on the isomer discrimination,and the improvement of analytical method of BAs. Further,this method was subsequently applied to elucidate in depth the chemical profile of BAs in yak bile. Regarding isomer discrimination for BAs,we constructed relative response-collision energy curves( RRCECs) by high performance liquid chromatographyion trap-time of flight-mass spectrometry( HPLC-IT-TOF-MS) in combination with high performance liquid chromatography-triple quadrupole-linear ion trap mass spectrometry( HPLC-Qtrap-MS). As a result,both the optimum collision energy( OCE) and CE_(50) exhibited great correlations with structural characteristics,thus enabling the isomer distinguishing,such as unconjugated BAs,glycine-conjugated BAs,and taurine-conjugated BAs. According to information provided by mass spectrometry,the comparison of OCE and CE_(50),retention time matching,combined with reference substances and database retrieval,a total of 30 bile acid derivatives were observed and identified in yak bile. The newly developed method could serve as a feasible tool for the in-depth characterization of BAs in bile and biological samples.