Smooth Steering Research Articles

Divergent visual ecology of Drosophila species drives object-tracking strategies matched to landscape sparsity

Maintaining stable gaze while tracking moving objects is commonplace across animal taxa, yet how diverse ecological needs impact these processes is poorly understood. During flight, the fruit-eating fly Drosophila melanogaster maintains course by making smooth steering adjustments to fixate the image of the distant visual background on the retina, while executing body saccades to investigate nearby objects such as food sources. Cactophilic Drosophila mojavensis live where there is no canopy; rather, the flora forming visual "background" and "objects" are one and the same. We tested whether D.mojavensis have adapted their flight control strategies for a visually sparse landscape. We used a magnetic tether that allows free movement in the yaw axis. In response to a textured bar moving across a similarly textured stationary background, D.melanogaster fixates the background, thereby stabilizing gaze while integrating bar dynamics to trigger tracking saccades. By contrast, two mojavensis subspecies in the repleta subgroup and one species in the melanogaster subgroup steer to smoothly fixate the bar, seemingly ignoring the stationary surround. Desert flies execute frequent bar-tracking saccades, but theirs are triggered when rotational velocity lags the bar. Thus, D.melanogaster, which lives in visually cluttered cosmopolitan habitats, leverages the optical disparities between nearby objects and distant foliage for a hybrid control strategy: "ground-fixate, object-saccade." Flies in distant phylogenetic subgroups with similar visual ecology use a "fixate-and-saccade" strategy, which would be adaptive in a visually sparse environment where individual landscape features are both approached and used to maintain a straight course.

Precise Positioning of Redundant Robot Motion Energy-Saving Control Based on Sensing Technology

In order to provide intelligent drinking water service for human beings, an intelligent indoor positioning water delivery system is proposed. The system can realize the functions of automatic filling water and so on. Based on the UP Squared processor, the system extends the UWB positioning Zigbee wireless transmission gyroscope manipulator and other functional modules. Based on the UP Squared processor, the system extends the UWB positioning Zigbee wireless transmission gyroscope manipulator and other functional modules. MPU6050 module was used to sense the rotation angle, and PID algorithm was used to control the rotation of redundant robot. The position of water cup is sensed based on current magnetic effect and electromagnetic induction law. Ultrasonic module is used to detect the rising height of the outlet and the water level in the cup. Several parameters of the system, such as orientation, distance adjustment, and power consumption, and the whole process of water delivery are tested through experiments. The experimental results show that the designed robot has a measurement accuracy of 1 cm, a total of static power of 5.35 W, and a total of dynamic power of 13.6 W. Conclusion. The system has the characteristics of accurate positioning, fast and smooth steering, and low power consumption, which can meet the technical requirements of practical application in indoor intelligent positioning water delivery.

Gymkhana and pylon slalom driving training effects on the cerebellum structure

To develop a suitable automobile design as per each driver’s characteristics and state, it is important to understand the brain function in acquiring driving skills. Reportedly, the brain structures of professionals, such as athletes and musicians, and those who have received training in special skills, undergo changes with training. However, the development process of the brain in terms of acquiring driving skills has not yet been clarified. In this study, we evaluated the effects of driving training on the brain and observed an increase in the volume of the right cerebellum after short-term training (3 days). The right cerebellum is responsible for controlling the right hand and right foot, which are important for driving. Drivers train to control a vehicle smoothly at high speeds at gymkhana and pylon slalom courses, which are often used in motor sports. The brain structure was analyzed before and after training using magnetic resonance imaging. Voxel-based morphometry was used to assess possible structural changes. First, the lap times after training were clearly shortened and vehicle dynamics were more stable, indicating that the drivers’ skill level clearly improved. Second, brain structural analysis revealed a volumetric increase in the right cerebellum. The cerebellum is involved in the process of learning sensory motor skills, such as smooth steering and pedal operations, driving course shape, and vehicle size perception. These results suggest a new inner model for driving operation and support the hypothesis that motor learning affects the cerebellum during vehicle driving training.

Sensorless Combined Control of In-wheel BLAC Motors for Mobility Scooter

This paper presents a speed control for in-wheel brushless AC (BLAC) motors using a simple hall sensor to improve steering and driving performance of mobility scooters. To achieve smooth steering and motion controls, the sensor-sensorless combined estimation method is proposed. Furthermore, a modified current model reference adaptive system (MRAS) is also proposed to increase the dynamic response and estimation accuracy. The dead-time effect is directly compensated to improve the sensorless performance in the lower speed region. From standstill to the controllable minimum speed region, fast open-loop control method is applied to the traction motors. The vehicle velocity is then controlled using the speed information obtained from the proposed sensorless-sensor combined estimation scheme. To verify the effectiveness, a prototype mobility scooter was manufactured and the proposed control method was tested. Experiment results shows that the steering and motion of the vehicle are improved and the overall performance is increased compared to that of the conventional driving method.

A Vehicle-Independent Autonomous Lane Keeping and Path Tracking System

This paper describes a lateral controller for path following developed to work on a wide range of vehicles without any changes in tuning between platforms. This vehicle independent path following controller employs a cascaded adaptive architecture, with the outer loop controlling the path dynamics and the inner loop controlling the vehicle dynamics. The controller is validated on four vehicle models in simulation at a variety of speeds. The controller is then implemented and validated in real-time on a Lincoln MKZ, a Class-8 Peterbilt 579 cab, a Peterbilt 579 with a loaded trailer, and a 1/10 th scale RC car. The controller is shown to perform well on the four simulated vehicles and the four experimental platforms at a range of longitudinal speeds, matching or exceeding human performance with consistent and smooth steering.

Development of crawler steering gearbox for combine harvester straight forward and steering in situ

Soil compaction and accumulation are caused by crawler steering. Soil characteristics would affect the ability of crawler straight forward and steering in situ. This paper studied the principle of straight forward steering and steering in situ mode of traditional unilateral brake steering gearbox of a crawler combine harvester. Then a positive and negative steering gearbox model was developed and processed. The performance of positive and negative steering gearbox was analyzed by performing two steering modes on soil compaction effect. Results showed that the positive and negative steering gearboxes can achieve three operating modes, such as straight forward, unilateral brake steering, and positive and negative steering in situ. The positive and negative steering in situ was smooth and without soil accumulation phenomenon. When the crawlers center distance, crawler grounding length, crawler width were 1150 mm, 1620 mm and 450 mm respectively, the rolling area and the scratch free area of positive and negative steering gearbox were 5.37 m2 and 0.36 m2 respectively. However, the rolling area and the scratch free area produced by unilateral brake steering gearbox were 13.92 m2 and 1.88 m2 respectively, which was 8.55 m2 larger than that of positive and negative steering gearbox. The crawler with positive and negative steering gearbox has minor damage to paddy soil. This research can provide a theoretical basis for the smooth steering in situ and structural optimization of the crawler steering gearbox. Keywords: steering gearbox, rolling soil track, steering in situ, rolling and scraping, shaft and gear DOI: 10.25165/j.ijabe.20201301.3957 Citation: Tang Z, Zhang H T, Li H C, Li Y M, Ding Z, Chen J S. Developments of crawler steering gearbox for combine harvester straight forward and steering in situ. Int J Agric & Biol Eng, 2020; 13(1): 120–126.

Developments of crawler steering gearbox for combine harvester straight forward and steering in situ

Soil compaction and accumulation are caused by crawler steering. Soil characteristics would affect the ability of crawler straight forward and steering in situ. This paper studied the principle of straight forward steering and steering in situ mode of traditional unilateral brake steering gearbox of a crawler combine harvester. Then a positive and negative steering gearbox model was developed and processed. The performance of positive and negative steering gearbox was analyzed by performing two steering modes on soil compaction effect. Results showed that the positive and negative steering gearboxes can achieve three operating modes, such as straight forward, unilateral brake steering, and positive and negative steering in situ. The positive and negative steering in situ was smooth and without soil accumulation phenomenon. When the crawlers center distance, crawler grounding length, crawler width were 1150 mm, 1620 mm and 450 mm respectively, the rolling area and the scratch free area of positive and negative steering gearbox were 5.37 m2 and 0.36 m2 respectively. However, the rolling area and the scratch free area produced by unilateral brake steering gearbox were 13.92 m2 and 1.88 m2 respectively, which was 8.55 m2 larger than that of positive and negative steering gearbox. The crawler with positive and negative steering gearbox has minor damage to paddy soil. This research can provide a theoretical basis for the smooth steering in situ and structural optimization of the crawler steering gearbox. Keywords: steering gearbox, rolling soil track, steering in situ, rolling and scraping, shaft and gear DOI: 10.25165/j.ijabe.20201301.3957 Citation: Tang Z, Zhang H T, Li H C, Li Y M, Ding Z, Chen J S. Developments of crawler steering gearbox for combine harvester straight forward and steering in situ. Int J Agric & Biol Eng, 2020; 13(1): 120–126.

Cloning Safe Driving Behavior for Self-Driving Cars using Convolutional Neural Networks

Background: In this paper, a Convolutional Neural Network (CNN) to learn safe driving behavior and smooth steering manoeuvring, is proposed as an empowerment of autonomous driving technologies. The training data is collected from a front-facing camera and the steering commands issued by an experienced driver driving in traffic as well as urban roads. Methods: This data is then used to train the proposed CNN to facilitate what it is called “Behavioral Cloning”. The proposed Behavior Cloning CNN is named as “BCNet”, and its deep seventeen-layer architecture has been selected after extensive trials. The BCNet got trained using Adam’s optimization algorithm as a variant of the Stochastic Gradient Descent (SGD) technique. Results: The paper goes through the development and training process in details and shows the image processing pipeline harnessed in the development. Conclusion: The proposed approach proved successful in cloning the driving behavior embedded in the training data set after extensive simulations.

Safe-driving cloning by deep learning for autonomous cars

In this paper, a convolutional neural network (CNN) to learn safe driving behaviour and smooth steering manoeuvring is proposed as an empowerment of autonomous driving technologies. The training data is collected from a front-facing camera and the steering commands issued by an experienced driver driving in traffic as well as urban roads. This data is then used to train the proposed CNN to facilitate what it is called 'behavioural cloning'. The proposed behaviour cloning CNN is named as 'BCNet', and its deep 17-layer architecture has been selected after extensive trials. The BCNet got trained using Adam's optimisation algorithm as a variant of the stochastic gradient descent (SGD) technique. The paper goes through the development and training process in details and shows the image processing pipeline harnessed in the development. The proposed approach is proved successful in cloning the driving behaviour embedded in the training dataset after extensive simulations.

Safe-driving cloning by deep learning for autonomous cars

In this paper, a convolutional neural network (CNN) to learn safe driving behaviour and smooth steering manoeuvring is proposed as an empowerment of autonomous driving technologies. The training data is collected from a front-facing camera and the steering commands issued by an experienced driver driving in traffic as well as urban roads. This data is then used to train the proposed CNN to facilitate what it is called 'behavioural cloning'. The proposed behaviour cloning CNN is named as 'BCNet', and its deep 17-layer architecture has been selected after extensive trials. The BCNet got trained using Adam's optimisation algorithm as a variant of the stochastic gradient descent (SGD) technique. The paper goes through the development and training process in details and shows the image processing pipeline harnessed in the development. The proposed approach is proved successful in cloning the driving behaviour embedded in the training dataset after extensive simulations.

Modelling and Position Control of an Electric Power Steering System

This paper presents the modelling, control and analysis of an axle parallel electric power steering system used for autonomous driving. The purpose of the controller is to ensure accurate and robust following of desired trajectories of front steering angle as well as to deliver a smooth steering wheel movement. Firstly the steering model shows how all nonlinearities for the elastic elements, frictions and gear ratio are considered. This model is linearised for the controller design. Thereafter open loop behaviour of the model is examined to determine system resonances. Next a controller synthesis using a Linear Quadratic Integrator controller as well as pole placement is developed. The performance of these controllers is presented in simulation. These results show good tracking performance from the two approaches presented and supports the use of a state space controller design approach in dealing with the various requirement of steering angle control for autonomous driving.

Design of a range extension strategy for power decentralized fuel cell/battery electric vehicles

Range extended electric vehicle is proposed to improve the range anxiety that drivers have regarding the battery running out on electric vehicles. Current design has compensated for this shortcoming by cascading a gasoline/diesel generator with the battery to facilitate the range of an electric vehicle. Due to the zero-CO2 emission stipulations, utilizing fuel cells as generators raises concerns in society. This paper presents a fuzzy charging strategy for power decentralized fuel cell/battery electric vehicles. In comparison to the conventional switch control, the proposed approach plays the role of enhancing the battery's state of charge (SOC). This approach improves the quick loss problem of the system's SOC and thus can offer a longer driving range. Smooth steering experience and range extension are employed as the main indexes for development of the fuzzy rules, which are mainly based on the energy management in the urban driving model. Evaluation of the entire control system is performed by simulation, which demonstrates its feasibility and effectiveness.

A Novel Range-Extended Strategy for Fuel Cell/Battery Electric Vehicles.

The range-extended electric vehicle is proposed to improve the range anxiety drivers have of electric vehicles. Conventionally, a gasoline/diesel generator increases the range of an electric vehicle. Due to the zero-CO2 emission stipulations, utilizing fuel cells as generators raises concerns in society. This paper presents a novel charging strategy for fuel cell/battery electric vehicles. In comparison to the conventional switch control, a fuzzy control approach is employed to enhance the battery's state of charge (SOC). This approach improves the quick loss problem of the system's SOC and thus can achieve an extended driving range. Smooth steering experience and range extension are the main indexes for development of fuzzy rules, which are mainly based on the energy management in the urban driving model. Evaluation of the entire control system is performed by simulation, which demonstrates its effectiveness and feasibility.

A Navigation-Field-Based Semi-Autonomous Nonholonomic Vehicle-Parking Assistant

We present a closed-loop approach to solve the semi-autonomous nonholonomic vehicle-parking assistant problem. The method allows for a priori unknown and potentially high vehicle speeds and dynamic environment variations. The computationally low-cost, memory-efficient algorithm respects constraints on the steering angle and its first two derivatives, allowing for comfortable, safe, and smooth steering motions. A parking situation is captured by an arrangement of precalculated vector fields of feasible nonholonomic paths, by which the vehicle is controlled toward a desired goal configuration inside a parking bay. Using standard onboard environment perception sensors, the system adapts to changes in the environment and avoids collisions whenever possible, without aborting a parking procedure.

The design of a look-down feedback adaptive controller for the lateral control of front-wheel-steering autonomous highway vehicles

An adaptive control law and a distance rate observer were developed for the lateral control of autonomous vehicles. The control law and the observer were implemented on a passenger vehicle. The test results show that both tight lateral tracking control and smooth steering can be achieved without feed forward information. The developed control law is shown to be very robust to changes in the most critical vehicle control parameters.

Analysis of a double-heterostructure spatial-phase controller for diode-laser beam steering

A numerical method to obtain the variation of an electric field as a noneigenmode lightwave propagates in a waveguide is discussed. The method is basically a mode expansion of the field using a hypothetical oversized waveguide with a modified index distribution. The hypothetical guide has an additive index step at remote boundaries, so that it has many discrete eigenmodes and, at the same time, permits light to propagate in the same way as in the original waveguide. The beam-steering properties of a double-heterostructure carrier-injection-type spatial-phase controller is then analyzed using this method. Conditions on the deflector structure and on the input light for smooth beam steering are discussed on the basis of the analysis. >