Steering Behavior Research Articles

A novel end-to-end model for steering behavior prediction of autonomous ego-vehicles using spatial and temporal attention mechanism

In recent years, autonomous vehicles have attracted many researchers from academia and industry. Their efforts include object detection and tracking in the fields of computer vision, probabilistic-fusion and decision-making algorithms, etc. But most current methods that directly map from image pixels to steering behavior are not able to generate accurate results, especially under scenarios such as light and weather change drastically; this is largely due to the fact that these methods ignore the temporal relationship between frames. In this paper, we propose a novel end-to-end deep learning framework based on temporal and spatial attention mechanism, which aims to solve the problem of inaccurate vehicle steering angle prediction in complex environments and the difficulty of model interpretation. First, we use video sequence and historical steering angle sequence as inputs to the model, instead of just using a single frame as input. Second, we design a temporal attention mechanism to capture the long- and short-term memory in input visual information, and a spatial attention mechanism to capture key objects in the image and obtain their position information. This is achieved by inserting carefully-designed SE-Net, ConvLSTM and CNN layers into the appropriate layers of the network framework. Finally, we demonstrate the feasibility of the proposed model with public Comma2k19, with comparison to current advanced methods. Experimental results show that compared with state-of-the-art methods, the average absolute error (MAE) values of our model on the training set and testing set are reduced by 10.2% and 6.3%, respectively, and has more accurate steering prediction performance. In addition, we explain the trigger mechanism of steering behavior prediction by visualizing the spatial attention map and temporal attention score on Comma2k19 and Udacity datasets, which further demonstrates that the proposed model can learn human-like driving behavior.

Implementation of Steering Process For Labriform Swimming Robot Based on Differential Drive Principle

The aim of this study is to develop a swimming robot with good steering performance, in which the steering behavior is achieved by one degree of freedom (1-DOF) represented by a two concave-shaped pectoral fins. The steering mechanism adopted here based on differential drive principle. This principle is carried out by varying the right/left fins velocities. Different radii have been achieved with four different cases of velocities. The proposed design has been validated theoretically via Solidworks{\textregistered} platform and proved practically in a physical swimming pool. A minimum turning radius achieved is 0.40 of body length (BL).

Investigating the Effect of Social and Cultural Factors on Drivers in Malaysia: A Naturalistic Driving Study.

Road accidents are increasing every year in Malaysia, and it is always challenging to collect reliable pre-crash data in the transportation community. Existing studies relied on simulators, police crash reports, questionnaires, and surveys to study Malaysia’s drivers’ behavior. Researchers previously criticized such methods for being biased and unreliable. To fill in the literature gap, this study presents the first naturalistic driving study in Malaysia. Thirty drivers were recruited to drive an instrumented vehicle for 750 km while collecting continuous driving data. The data acquisition system consists of various sensors such as OBDII, lidar, ultrasonic sensors, IMU, and GPS. Irrelevant data were filtered, and experts helped identify safety criteria regarding multiple driving metrics such as maximum acceptable speed limits, safe accelerations, safe decelerations, acceptable distances to vehicles ahead, and safe steering behavior. These thresholds were used to investigate the influence of social and cultural factors on driving in Malaysia. The findings show statistically significant differences between drivers based on gender, age, and cultural background. There are also significant differences in the results for those who drove on weekends rather than weekdays. The study presents several recommendations to various public and governmental sectors to help prevent future accidents and improve traffic safety.

The coordination of gaze and steering behavior in drone-racing pilots during high-speed flight

The coordination of gaze and steering behavior in drone-racing pilots during high-speed flight

A unified mechanism for innate and learned visual landmark guidance in the insect central complex.

Insects can navigate efficiently in both novel and familiar environments, and this requires flexiblity in how they are guided by sensory cues. A prominent landmark, for example, can elicit strong innate behaviours (attraction or menotaxis) but can also be used, after learning, as a specific directional cue as part of a navigation memory. However, the mechanisms that allow both pathways to co-exist, interact or override each other are largely unknown. Here we propose a model for the behavioural integration of innate and learned guidance based on the neuroanatomy of the central complex (CX), adapted to control landmark guided behaviours. We consider a reward signal provided either by an innate attraction to landmarks or a long-term visual memory in the mushroom bodies (MB) that modulates the formation of a local vector memory in the CX. Using an operant strategy for a simulated agent exploring a simple world containing a single visual cue, we show how the generated short-term memory can support both innate and learned steering behaviour. In addition, we show how this architecture is consistent with the observed effects of unilateral MB lesions in ants that cause a reversion to innate behaviour. We suggest the formation of a directional memory in the CX can be interpreted as transforming rewarding (positive or negative) sensory signals into a mapping of the environment that describes the geometrical attractiveness (or repulsion). We discuss how this scheme might represent an ideal way to combine multisensory information gathered during the exploration of an environment and support optimal cue integration.

Neural-Fuzzy-Based Adaptive Sliding Mode Automatic Steering Control of Vision-based Unmanned Electric Vehicles

This paper presents a novel neural-fuzzy-based adaptive sliding mode automatic steering control strategy to improve the driving performance of vision-based unmanned electric vehicles with time-varying and uncertain parameters. Primarily, the kinematic and dynamic models which accurately express the steering behaviors of vehicles are constructed, and in which the relationship between the look-ahead time and vehicle velocity is revealed. Then, in order to overcome the external disturbances, parametric uncertainties and time-varying features of vehicles, a neural-fuzzy-based adaptive sliding mode automatic steering controller is proposed to supervise the lateral dynamic behavior of unmanned electric vehicles, which includes an equivalent control law and an adaptive variable structure control law. In this novel automatic steering control system of vehicles, a neural network system is utilized for approximating the switching control gain of variable structure control law, and a fuzzy inference system is presented to adjust the thickness of boundary layer in real-time. The stability of closed-loop neural-fuzzy-based adaptive sliding mode automatic steering control system is proven using the Lyapunov theory. Finally, the results illustrate that the presented control scheme has the excellent properties in term of error convergence and robustness.

Einstein–Podolsky–Rosen steering testing via quantum measurement

Einstein–Podolsky–Rosen (EPR) steering is a method that is used to analyze the physical inseparability of a quantum system state. Theoretically, the preferences of EPR steering behaviour in the inseparability detection can be suitably characterized by means of steering criterion. Here, we propose a scheme to test the EPR steering of a bipartite entangled state via quantum measurement. Based on the operation angles involved in the measurement, we derive a series of conditions that determine whether the EPR steering exists or not, and further prove that for the specific initial state the EPR steering can be tested. As an application, we apply our scheme to identify the steerability of the asymmetry state and Werner state, and the steering testing can be realized by adjusting the operation angles.

Real Time Driver Drowsiness Detection App using ML

Every year road accidents are increasing rapidly as technological and mechanical advancements in vehicles permits drivers to drive at high speed. Approximately 1.35 million people die each year as a result of road accidents in India alone 151 thousand casualties were recorded last year. From this nearly 78% road accidents are caused due to driver's fault. Main factors for this accidents are drowsiness, drunk and drive and over speeding from which nearly 40% of the accidents caused due to drowsiness. People are conscious about the risk of drinking and driving but don’t realize the dangerous of drowsiness because no instruments exist to measure the driver drowsiness. If the Driver fails to concentrate on driving it reduces the driver reaction time and impairs steering behavior To solve this problem we are going to use the power of machine learning to identify if the driver is drowsy or not. Generally when someone feels drowsy his\hers eye blinking speed decreases by specifying threshold value we can detect if the driver is drowsy or not. This programme provides a mechanism for scanning facial landmarks and then using the essential landmarks for eye tracking after recognising the face. this insures that the driver is in full control of his vehicle. The system make use of device’s front camera to monitor drivers’s face to detect drowsiness and alarms the driver if system founds him drowsy. More functionality will be added to the system, such as issuing an SOS if something occurs to a car in a remote location. The software also has emergency numbers, so that in the event of an emergency, the driver may call the appropriate authorities as needed. This system may also be used for navigating by utilising the app's map functionality. Flutter will be utilised to provide a native and user-friendly system interface. As a result, the software will be available for both Android and iOS smartphones. The use of as little hardware as possible will ensure seamless processing. This technology may be employed in a variety of circumstances, including cab services, state transportation and on-road goods delivery.

Multi‐beam steered harmonic pattern synthesis in timed antenna array with optimized and pre‐defined RF switching

AbstractOptimal harmonic beam steering of the time‐modulated linear antenna array (TMLAA) with optimized and pre‐defined ON and OFF switching schemes are proposed in this work. Three different approaches are considered to obtain various beam steering behaviors of TMLAA. The first approach is considered for the optimization of ON–OFF switching sequence for the steering of the harmonic beams toward approximately ±30° along with the simultaneous reduction of side lobe level (SLL) and side band level (SBL). Second approach is considered for the optimization of ON–OFF switching sequence for the steering of the harmonic beams toward approximately ±30° and ±60° simultaneously. Third approach is considered for the optimal reduction of SLL and SBL of harmonically steered beams for a pre‐defined ON–OFF switching scheme. Two unique cost function is proposed in this work to achieve the optimal parameter for above mentioned three different approaches. The first cost function is used for the optimization of Ton and Toff switching sequence and antenna element spacing of the array. These optimal values will lead to steer the main beam of first positive and negative harmonic radiation from the main beam of the fundamental radiation pattern and reduce SLL and SBL simultaneously. Second cost function deals with the optimization of excitation amplitude and distance between any two consecutive antennas of the array. These optimal values lead to reducing the SLL and SBL simultaneously harmonically steered beam obtained by pre‐defined Ton and Toff switching sequence. Optimization algorithm applied to get the optimal value of the variable parameters is Novel Particle Swarm Optimization. Further to explain the stated proposed approach, four different sets of 10‐, 16‐, 20‐, 32‐element TMLAA are taken as an example. Some of the simulation results are validated through electromagnetic simulator to validate the result obtained by MATLAB.

Vehicle handling improvements through Steer-by-Wire

This paper focuses on handling improvements enabled through Steer-by-Wire systems, which have increasingly become subject of R&D, as they not only offer the potential for improving vehicle handling but also have many advantages in combination with automated driving. Handling improvements through a steering ratio depending on vehicle speed, as well as steering-wheel angle, are known from Active Front Steering systems. A new overall concept is proposed, that also takes into account lateral and longitudinal acceleration as well as steering rate, which are all available signals in a production car. The overall concept is designed in an optimization process to modify a range of established characteristic parameters known from open-loop maneuvers and the objective evaluation of vehicle handling. In this context, validated models for a vehicle and a Steer-by-Wire system are used to obtain reliable results in simulation. Possibilities for tuning the non-linear steering behavior as well as improvements in the dynamic behavior, especially in yaw damping and response time, are demonstrated.

Influences of wing gate turnstiles' characteristics on pedestrian evacuation based on agent-based egress model.

Wing gate turnstiles are widely used in airports, subway stations, railway stations, movie theaters, parks, etc., and influence the pedestrian evacuation greatly while accidents happen. Thus, doing some researches on how to improve the evacuation speed for wing gate turnstiles are meaningful and necessary. In response to this issue, the influences of the wing gate turnstiles on pedestrian evacuation are investigated by pathfinder in this paper. Pathfinder is an agent-based egress simulator that uses steering behaviors to model occupant motion. First of all, by Pathfinder, two evacuation models with flat-shaped and wedge-shaped hosts wing gate turnstiles, respectively, are established, and the simulations of pedestrian evacuation are carried out by computer. The results show the evacuation speed obtained by the wedge-shaped host is faster than that obtained by the flat-shaped host. Moreover, the longest evacuation time is obtained when the channels' width is 90 cm, and the shortest is obtained when the channels' width is 51.4 cm. For a small number of evacuees, the evacuation time is mainly influenced by the original distributions of the evacuees, however, as the number of the evacuees increases, the evacuation time is mainly influenced by the number of the evacuees. Furthermore, the wing gate turnstiles with widened hosts can get less evacuation time no matter whether the hosts are flat-shaped or wedge-shaped, however, the evacuation-time-saved rate obtained by wedge-shaped host is much higher than that obtained by flat-shaped host. Then, the influences of the wedge angle on the pedestrian evacuation are discussed, and the optimal wedge angle is obtained. The obtained results can not only provide helps for managers to evacuate crowds under emergency, but also offer some assistances for designers to design those gate turnstiles or building exits.

Two Nash-equilibrium-based steering control models for representing a driver’s interaction with vehicle automated steering

ABSTRACT Automated steering technology offers significant benefits to the safety of vehicles, but desire to keep the human driver in the loop requires a better understanding of the interaction between driver and vehicle. An existing noncooperative–game–theoretic framework for modelling such interaction is revisited, leading to the development of two driver steering control models. Both bear Nash–equilibrium properties, but involve different assumptions about driver steering behaviour. A simulation study is performed to demonstrate the difference between the two driver models. An experiment using a fixed–base driving simulator is conducted to measure six test subjects' steering angles in response to the lane–change manoeuvres generated by an automated steering controller. The two driver models' capabilities for representing driver steering behaviour are investigated through fitting them to measured driver steering angles. Key model parameters are identified using a system identification procedure. It is found that the two driver models have equivalent capability in capturing the trend of the six test subjects' measured steering angles, but less good at reproducing the overshoot and oscillation involved in two subjects' steering angles. It is found that the inclusion of an arm neuromuscular system model can improve the performance of the proposed driver models.

Extraction of optimal fatigue‐driving steering indicators considering individual differences

Extracting effective steering indicators is critical for fatigue-identification based on steering behaviour. However, considering individual differences, the best calculation parameters of the individual driver's steering indicators are different. The indicator's performance decreases by using unified parameters rather than the individual driver's best parameters. The authors propose a model extracting individual driver's optimal fatigue steering indicators calculated by the individual driver's best parameters. Individual driver's naturalistic driving data are analysed by the model. First, indicators in sober and fatigue state are examined by the Wilcoxon test, and |Z| represents the indicator's fatigue-identification performance. And the function reflecting correspondences between calculation parameters and indicator's fatigue-identification performance is constructed. Then the function is optimized using particle swarm optimization to obtain individual driver's best calculation parameters maximizing |Z|. Finally, effective indicators calculated by the individual driver's best parameters constitute the individual driver's optimal fatigue indicator set. Indicators calculated by the individual driver's best parameters and unified parameters are used to establish the individual fatigue-identification model. The average identification accuracy of models using individual driver's best parameters and unified parameters is 87.4% and 79.1%, which indicates using individual driver's optimal fatigue indicators can improve identification accuracy. This study can provide references for individualized fatigue-driving identification

Analytical Study on the Cornering Behavior of an Articulated Tracked Vehicle

Articulated tracked vehicles have been traditionally studied and appreciated for the extreme maneuverability and mobility flexibility in terms of grade and side slope capabilities. The articulation joint represents an attractive and advantageous solution, if compared to the traditional skid steering operation, by avoiding any trust adjustment between the outside and inside tracks. This paper focuses on the analysis and control of an articulated tracked vehicle characterized by two units connected through a mechanical multiaxial joint that is hydraulically actuated to allow the articulated steering operation. A realistic eight degrees of freedom mathematical model is introduced to include the main nonlinearities involved in the articulated steering behavior. A linearized vehicle model is further proposed to analytically characterize the cornering steady-state and transient behaviors for small lateral accelerations. Finally, a hitch angle controller is designed by proposing a torque-based and a speed-based Proportional Integral Derivative (PID) logics. The controller is also verified by simulating maneuvers typically adopted for handling analysis.

Persistent thermal input controls steering behavior in Caenorhabditis elegans.

Motile organisms actively detect environmental signals and migrate to a preferable environment. Especially, small animals convert subtle spatial difference in sensory input into orientation behavioral output for directly steering toward a destination, but the neural mechanisms underlying steering behavior remain elusive. Here, we analyze a C. elegans thermotactic behavior in which a small number of neurons are shown to mediate steering toward a destination temperature. We construct a neuroanatomical model and use an evolutionary algorithm to find configurations of the model that reproduce empirical thermotactic behavior. We find that, in all the evolved models, steering curvature are modulated by temporally persistent thermal signals sensed beyond the time scale of sinusoidal locomotion of C. elegans. Persistent rise in temperature decreases steering curvature resulting in straight movement of model worms, whereas fall in temperature increases curvature resulting in crooked movement. This relation between temperature change and steering curvature reproduces the empirical thermotactic migration up thermal gradients and steering bias toward higher temperature. Further, spectrum decomposition of neural activities in model worms show that thermal signals are transmitted from a sensory neuron to motor neurons on the longer time scale than sinusoidal locomotion of C. elegans. Our results suggest that employments of temporally persistent sensory signals enable small animals to steer toward a destination in natural environment with variable, noisy, and subtle cues.

DCFS-based deep learning supervisory control for modeling lane keeping of expert drivers

In this paper, a novel driver model for lane keeping is proposed to replicate the steering behavior of expert drivers. Specifically, a feedforward-feedback control scheme mocking expert drivers is adopted: the feedforward controller plays a leading role, which is a data-driven model based on deep convolutional fuzzy systems (DCFS), and for the sake of human-simulation and guaranteed stability, a supervisory feedback controller is designed, which works only if the state hits the set boundary. Comparing with the previous driver models, the key novelty of the paper is to introduce the “motor intermittency” of human behavior into driver modeling, which is an important issue for biological modeling of the drivers. Simulations on the joint platform of PreScan and CarSim show that the newly presented driver model has better matching performance to the expert drivers comparing with the two different types of advanced model predictive control (MPC) controllers. The proposed driver model has the potential application for semi-automated vehicles to provide human-like qualities for automated driving, which may be one of the essential points to promote the comfort when the driver hands over the steering authority, and improve the transition smoothness in the scenario of human vehicle co-piloting.

A Machine Learning-Based Approach to Analyze Information Used for Steering Control

Understanding the relationship between driving behavior and visual information is important for holistic understanding of driving behavior. However, the analysis of the cognitive behavior for steering/throttle control has been only conducted under a special simulator environment. Therefore, in this study, we aimed to develop a convolutional neural network (CNN) with human physical characteristics to analyze the driver's cognitive behavior and to validate that the machine learning methods can be an analytical method for understanding driver behavior. We obtained the driving data in a simulator experiment to train the proposed CNN model. The region where the visual field influences drivers' steering behavior was analyzed using the results of the feature maps generated by the trained CNN model and the driver's gaze behavior. The results indicate that the driver performs steering control using the information within 20 degrees from the gaze point. This shows that the results obtained from our proposed method can reproduce the same results as previous findings. We also validated that the results are not uniquely obtained depending on the proposed model and environment but are also influenced by the driving behavior such as the gaze point and the steering control. We analyzed the dataset generated by the mathematical control model, called the driver model, which performs different behaviors from the driver. The analysis results generated by the driver model were different from the results of the human data. Therefore, the results generated by the machine learning-based analysis are influenced by the driving behavior. Consequently, these results imply that machine learning methods have the potential to become analytical methods for understanding driver behavior.

Pigeon Robot for Navigation Guided by Remote Control: System Construction and Functional Verification

Animal robots have outstanding advantages over traditional robots in their own energy supplies, orientation, and natural concealment, delivering significant value in the theories and applications of neural science, national security, and other fields. Presently, many animal robots have been fabricated, but researches about the applications of avian robots are still lacking. In this study, we constructed a Pigeon Robot System (PRS), optimized the electric stimulation parameters, assessed the electric stimulus of navigation, and evaluated the navigation efficiency in the field. Biphasic pulse constant current pattern was adapted, and the optimal stimulus parameters of 4 nuclei tested were of amplitude 0.3 mA, 5 pulse trains, frequency 25 Hz, 5 pulses, and a 25% duty cycle. Effective ratio of left and right steering behavior response to electric stimulus dorsointermedius ventralis anterior nuclei was 67% and 83%, respectively (mean value 75%). Electrical stimulation efficiency was 0.34–0.68 and path efficiency was 0.72–0.85 among pigeon robot individuals in the open field. Neither electrical stimulation efficiency nor path efficiency differed significantly (P > 0.05), suggesting that the navigational PRS performance was not biased in either direction. PRS can achieve continuous navigation along simple pathways and provide the necessary application infrastructure and technical reference for the development of animal robot navigation technology.

Theoretical and experimental investigation of driver noncooperative-game steering control behavior

This paper investigates two noncooperative-game strategies which may be used to represent a human driver's steering control behavior in response to vehicle automated steering intervention. The first strategy, namely the Nash strategy is derived based on the assumption that a Nash equilibrium is reached in a noncooperative game of vehicle path-following control involving a driver and a vehicle automated steering controller. The second one, namely the Stackelberg strategy is derived based on the assumption that a Stackelberg equilibrium is reached in a similar context. A simulation study is performed to study the differences between the two proposed noncooperative-game strategies. An experiment using a fixed-base driving simulator is carried out to measure six test driver's steering behavior in response to vehicle automated steering intervention. The Nash strategy is then fitted to measured driver steering wheel angles following a model identification procedure. Control weight parameters involved in the Nash strategy are identified. It is found that the proposed Nash strategy with the identified control weights is capable of representing the trend of measured driver steering behavior and vehicle lateral responses. It is also found that the proposed Nash strategy is superior to the classic driver steering control strategy which has widely been used for modeling driver steering control over the past. A discussion on improving automated steering control using the gained knowledge of driver noncooperative-game steering control behavior was made.

On the nature of eye-hand coordination in natural steering behavior.

Eyes and hand movements are known to be coordinated during a variety of tasks. While steering a vehicle, gaze was observed to be tightly linked with steering wheel angle changes over time, with the eyes leading the hands. In this experiment, participants were asked to drive a winding road composed of bends with systematically manipulated radii of curvature, under regular and automatic steering conditions. With automatic steering, the vehicle followed the road, but the steering wheel and participants hands did not move. Despite the absence of physical eye-hand coordination in that condition, the eye and [what the hands should have done] to produce the action on the steering wheel were found to be coordinated, as under regular steering. This result brings a convincing piece of evidence that eye movements do more than just guiding the hands. In addition, eye-hand coordination was also found to be intermittent, context and person-dependant.