Autonomous Land Vehicle Research Articles

Novel EKF-Based Vision/Inertial System Integration for Improved Navigation

With advances in computing power, stereo vision has become an essential part of navigation applications. However, there may be instances wherein insufficient image data precludes the estimation of navigation parameters. Earlier, a novel vision-based velocity estimation method was developed by the authors, which suffered from the aforementioned drawback. In this paper, the vision-based navigation method has been integrated with a unique low-cost reduced inertial sensor system to bridge the navigation gap using one gyroscope and two accelerometers along with the inputs from wheel speed sensors. The integrated system is based on the extended Kalman filter and was tested on three trajectories with the introduction of vision data gaps. The system showed promising results for autonomous land vehicle applications.

An Efficient Scan-to-Map Matching Approach Based on Multi-channel Lidar

Accurately localizing the vehicle against a pre-built high precision map is an essential step for the Autonomous Land Vehicle (ALV). This paper proposes an efficient scan-to-map matching approach based on multi-channel lidar. We firstly advocate the usage of both the lidar reflectance map and the height map, as these two maps contain complementary information. Then, borrowing ideas from the Lucas-Kanade optical flow approach, we formulate the scan-to-map matching problem in a similar form, and propose an efficient gradient descent approach to solve it. Finally, the proposed approach is integrated into a filtering framework for real-time online localization. Experiments on real-world dataset have demonstrated the validity of our approach.

Counter-cyclical public venture capital: Debt-funding as an anti-austerity innovation strategy

This article outlines a counter-cyclical innovation strategy to achieve prosperity, derived from an innovative project, the California Institute for Regenerative Medicine (CIRM). We identify an ‘innovation paradox’ in that the very point in the business cycle, when legislators are tempted to view austerity as a cure for economic downturns and to reduce innovation spend, is when an increase is most needed to create new industries and jobs and innovate out of recession or depression. It is both desirable and possible that policymakers resist the urge to capitulate to the innovation paradox. During periods that exhibit subdued inflation, elevated spare productive capacity, and low government borrowing rates, governments should increase their borrowings and use the proceeds to boost investment targeted towards innovation. We show how the State of California successfully utilized debt financing, traditionally reserved for physical infrastructure projects, to stimulate the development of intellectual infrastructure. Finally, we recommend a halt to European austerity policies and a ‘triple helix’ broadening of narrow ‘smart specialization’ policies that chase a private venture capital chimera. Europe should seize the present macroeconomic opportunity of low interest rates, borrow for innovation and be paid back manifold by ‘picking winners’, similarly to what the USA has been doing through DARPA (Defense Advanced Research Projects Agency) with GPS, as a response to Sputnik, the Internet and artificial intelligence, or the driverless car, formerly known as the ‘autonomous land vehicle’ in its military guise. Proactively targeted macroscopic investments in innovation are needed to solve the productivity/employment puzzle and foster the transition to a knowledge-based society.

A feature matching and fusion-based positive obstacle detection algorithm for field autonomous land vehicles

Positive obstacles will cause damage to field robotics during traveling in field. Field autonomous land vehicle is a typical field robotic. This article presents a feature matching and fusion-based algorithm to detect obstacles using LiDARs for field autonomous land vehicles. There are three main contributions: (1) A novel setup method of compact LiDAR is introduced. This method improved the LiDAR data density and reduced the blind region of the LiDAR sensor. (2) A mathematical model is deduced under this new setup method. The ideal scan line is generated by using the deduced mathematical model. (3) Based on the proposed mathematical model, a feature matching and fusion (FMAF)-based algorithm is presented in this article, which is employed to detect obstacles. Experimental results show that the performance of the proposed algorithm is robust and stable, and the computing time is reduced by an order of two magnitudes by comparing with other exited algorithms. This algorithm has been perfectly applied to our autonomous land vehicle, which has won the champion in the challenge of Chinese “Overcome Danger 2014” ground unmanned vehicle.

Likelihood-Field-Model-Based Dynamic Vehicle Detection and Tracking for Self-Driving

Dynamic vehicle detection and tracking is crucial for self-driving in urban environments. The main problem of the previous beam-model-based algorithms is that they cannot detect and track dynamic vehicles that are occluded by other objects. In this paper, we develop a novel dynamic vehicle detection and tracking algorithm to solve this problem for our autonomous land vehicle (ALV), which is equipped with a Velodyne LIDAR and a GPS-aid inertial navigation system. For detection, our improved two-dimensional virtual scan is presented to detect the potential dynamic vehicles with a scan differencing operation. Then, for each potential dynamic vehicle, a novel likelihood-field-based vehicle measurement model is proposed to weight its possible poses. Finally, our newly modified scaling series algorithm and the importance sampling technique are adopted to estimate the initial pose and the corresponding velocity for each vehicle, respectively. The scaling series algorithm coupled with a Bayesian filter (SSBF) was previously used to handle the tactile localization problem in static background scenes. For tracking dynamic vehicles, we improve the SSBF by adding the ego-motion compensation so that the improved algorithm is able to update the pose and velocity for each vehicle in dynamic background scenes. Both the quantitative and qualitative experimental results validate the performance of our dynamic vehicle detection and tracking algorithm on the KITTI datasets and the Velodyne data collected by our ALV in dynamic urban environments.

A dynamic filtering obstacle detection method based on ultrasonic array

In research on the navigation and control of an Autonomous Land Vehicle (ALV), the ultrasonic obstacle detection system plays an important role extending the environment cognition capability of the ALV. With a goal of improving the accuracy of ultrasonic obstacle detection, a dynamic data filtering method based on ultrasonic array is presented. The sonar return data is first processed through static filtering according to the geometric relationship of the ultrasonic array. Then, dynamic filtering is executed using the orientation and trajectory information of the vehicle. The dynamic filtering method is compared to the traditional ultrasonic obstacle detection method, which is the static filtering method in a typical field environment. The experiment result demonstrates the validity of the method for solving the fake data problem, and the accuracy of obstacle detection is improved.

The role of robots in the battlefields of the future

Purpose This paper aims to provide an overview of robots presently in use by the military and an insight into some that are under development. Design/methodology/approach Following a short introduction, this paper first considers existing applications of robots in the military field, including details of Russian weaponised ground robots. It then highlights a range of military robot developments and concludes with a brief discussion. Findings Drones (unmanned aerial vehicles) and small unmanned ground vehicles (UGVs) are among the most widely used robots by the military. Russia is developing a growing armoury of heavily weaponised UGVs, some of which were recently deployed in Syria. Some topics of development include humanoid robots, powered exoskeletons, load-carrying robots, micro-air vehicles and autonomous land vehicles. Robots will play an ever-growing role in military actions, and while some developments offer longer-term prospects, others are expected to be deployed in the near future. Originality/value Robots are playing an increasingly important role in military conflicts, and this provides details of present-day and anticipated future uses of robots by the military.

Lateral Path Tracking Control of Autonomous Land Vehicle Based on ADRC and Differential Flatness

This paper presents a tracking control method for the lateral motion of an autonomous land vehicle (ALV). This method is based on active disturbance rejection control (ADRC) scheme and differential flatness theory. The lateral motion is hard to control since it is underactuated, nonlinear, and with large uncertainties. By making a small-angle approximation, the dynamic model is linearized. The flatness of the linear model is proved and a flat output is found. An equivalent form of the model is obtained based on the flat output and its derivatives only. Moreover, an ADRC is adopted to guarantee both control accuracy and strong robustness. Simulation results are presented and the results show the effectiveness of the control strategies.

Location aware sensor routing protocol for mobile wireless sensor networks

Location aware sensor routing (LASeR) protocol is a novel solution to the challenges of routing in mobile wireless sensor networks (MWSNs). It addresses the high reliability and low latency requirements of emerging applications. The protocol uses location information to maintain a gradient field even in highly mobile environments, whilst reducing the routing overhead. This allows the protocol to utilise a blind forwarding technique to propagate packets towards the sink. The protocol inherently utilises multiple paths simultaneously to create route diversity and increase its robustness. LASeR is designed for use in a high variety of MWSN applications with autonomous land, sea or air vehicles. Analytical expressions are derived and evaluated against the simulations. Extensive modelling and simulation of the proposed routing protocol has shown it to be highly adaptable and robust. It is compared with the recent MWSN proactive highly ambulatory sensor routing protocol, the high performance mobility adaptive cross-layer routing protocol, as well as ad-hoc on-demand distance vector and optimised link state routing. Protocols are evaluated on packet delivery ratio, end-to-end delay, overhead, throughput and energy consumption. The results highlight both the high performance of LASeR in various challenging environments and its superiority over the state-of-the-art.

Build and Optimization of an Intelligent Decision Support System for UGVs Based on the Human-Robot Coordination

Unmanned systems of autonomy is a kind of perception, observation, analysis, communication, planning, decision-making and action ability, and complete human tasks assigned by the human-computer interaction it. Ground unmanned systems or in academic research field known as mobile robot autonomous land vehicle, known as the unmanned ground vehicles in the army. Based on variable independently, fault diagnosis, situation assessment, decision support, on the basis of analysis and research, building contains a variety of human-computer interaction in the form of unmanned ground vehicles, the intelligent decision support system by determining hierarchy of human-computer interaction, so as to achieve optimal human- robot collaborative, lay a certain foundation for the research of unmanned ground vehicle autonomy.

Monocular Road Detection Using Structured Random Forest

Road detection is a key task for autonomous land vehicles. Monocular vision-based road-detection algorithms are mostly based on machine learning approaches and are usually cast as classification problems. However, the pixel-wise classifiers are faced with the ambiguity caused by changes in road appearance, illumination and weather. An effective way to reduce the ambiguity is to model the contextual information with structured learning and prediction. Currently, the widely used structured prediction model in road detection is the Markov random field or conditional random field. However, the random field-based methods require additional complex optimization after pixel-wise classification, making them unsuitable for real-time applications. In this paper, we present a structured random forest-based road-detection algorithm which is capable of modelling the contextual information efficiently. By mapping the structured label space to a discrete label space, the test function of each split node can be trained in a similar way to that of the classical random forests. Structured random forests make use of the contextual information of image patches as well as the structural information of the labels to get more consistent results. Besides this benefit, by predicting a batch of pixels in a single classification, the structured random forest-based road detection can be much more efficient than the conventional pixel-wise random forest. Experimental results tested on the KITTI-ROAD dataset and data collected in typical unstructured environments show that structured random forest-based road detection outperforms the classical pixel-wise random forest both in accuracy and efficiency.

LiDAR Based Negative Obstacle Detection for Field Autonomous Land Vehicles

Negative obstacles for field autonomous land vehicles ALVs refer to ditches, pits, or terrain with a negative slope, which will bring risks to vehicles in travel. This paper presents a feature fusion based algorithm FFA for negative obstacle detection with LiDAR sensors. The main contributions of this paper are fourfold: 1 A novel three-dimensional 3-D LiDAR setup is presented. With this setup, the blind area around the vehicle is greatly reduced, and the density of LiDAR data is greatly improved, which are critical for ALVs. 2 On the basis of the proposed setup, a mathematical model of the point distribution of a single scan line is deduced, which is used to generate ideal scan lines. 3 With the mathematical model, an adaptive matching filter based algorithm AMFA is presented to implement negative obstacle detection. Features of simulated obstacles in each scan line are employed to detect the real negative obstacles. They are supposed to match with features of the potential real obstacles. 4 Grounded on AMFA algorithm, a feature fusion based algorithm is proposed. FFA algorithm fuses all the features generated by different LiDARs or captured at different frames. Bayesian rule is adopted to estimate the weight of each feature. Experimental results show that the performance of the proposed algorithm is robust and stable. Compared with the state-of-the-art techniques, the detection range is improved by 20%, and the computing time is reduced by an order of two magnitudes. The proposed algorithm had been successfully applied on two ALVs, which won the champion and the runner-up in the Overcome Danger 2014 ground unmanned vehicle challenge of China.

Robust curb detection with fusion of 3D-Lidar and camera data.

Curb detection is an essential component of Autonomous Land Vehicles (ALV), especially important for safe driving in urban environments. In this paper, we propose a fusion-based curb detection method through exploiting 3D-Lidar and camera data. More specifically, we first fuse the sparse 3D-Lidar points and high-resolution camera images together to recover a dense depth image of the captured scene. Based on the recovered dense depth image, we propose a filter-based method to estimate the normal direction within the image. Then, by using the multi-scale normal patterns based on the curb's geometric property, curb point features fitting the patterns are detected in the normal image row by row. After that, we construct a Markov Chain to model the consistency of curb points which utilizes the continuous property of the curb, and thus the optimal curb path which links the curb points together can be efficiently estimated by dynamic programming. Finally, we perform post-processing operations to filter the outliers, parameterize the curbs and give the confidence scores on the detected curbs. Extensive evaluations clearly show that our proposed method can detect curbs with strong robustness at real-time speed for both static and dynamic scenes.

Robust Zebra-Crossing Detection for Autonomous Land Vehicles and Driving Assistance Systems

Road scene understanding is critical for driving assistance systems and autonomous land vehicles. The main function of road scene understanding is robustly detecting useful visual objects existing in a road scene. A zebra crossing is a typical pedestrian crossing used in many countries around the world. When detecting a zebra crossing, an autonomous lane vehicle is normally required to automatically slow down its speed and to trigger a path-planning strategy for passing the zebra crossing. Also, most of driving assistance systems can send an early-warning signal to remind drivers to be more careful. This paper proposes a robust zebra-crossing detection algorithm for autonomous land vehicles and driving assistance systems. Firstly, an inverse perspective map is generated by utilizing camera calibration parameters to obtain a bird-eye view road image. Secondly, a course-to-fine detection process is applied to obtain a candidate zebra-crossing region and finally a true zebra-crossing region is recognized by combining appearance and shape features. Experiments on several kinds of real road videos which also include several challenge scenes demonstrate the effectiveness and efficiency of the proposed method.

Self-Learning Cruise Control Using Kernel-Based Least Squares Policy Iteration

This paper presents a novel learning-based cruise controller for autonomous land vehicles (ALVs) with unknown dynamics and external disturbances. The learning controller consists of a time-varying proportional-integral (PI) module and an actor-critic learning control module with kernel machines. The learning objective for the cruise control is to make the vehicle's longitudinal velocity follow a smoothed spline-based speed profile with the smallest possible errors. The parameters in the PI module are adaptively tuned based on the vehicle's state and the action policy of the learning control module. Based on the state transition data of the vehicle controlled by various initial policies, the action policy of the learning control module is optimized by kernel-based least squares policy iteration (KLSPI) in an offline way. The effectiveness of the proposed controller was tested on an ALV platform during long-distance driving in urban traffic and autonomous driving on off-road terrain. The experimental results of the cruise control show that the learning control method can realize data-driven controller design and optimization based on KLSPI and that the controller's performance is adaptive to different road conditions.

Gaussian-Process-Based Real-Time Ground Segmentation for Autonomous Land Vehicles

Ground segmentation is a key component for Autonomous Land Vehicle (ALV) navigation in an outdoor environment. This paper presents a novel algorithm for real-time segmenting three-dimensional scans of various terrains. An individual terrain scan is represented as a circular polar grid map that is divided into a number of segments. A one-dimensional Gaussian Process (GP) regression with a non-stationary covariance function is used to distinguish the ground points or obstacles in each segment. The proposed approach splits a large-scale ground segmentation problem into many simple GP regression problems with lower complexity, and can then get a real-time performance while yielding acceptable ground segmentation results. In order to verify the effectiveness of our approach, experiments have been carried out both on a public dataset and the data collected by our own ALV in different outdoor scenes. Our approach has been compared with two previous ground segmentation techniques. The results show that our approach can get a better trade-off between computational time and accuracy. Thus, it can lead to successive object classification and local path planning in real time. Our approach has been successfully applied to our ALV, which won the championship in the 2011 Chinese Future Challenge in the city of Ordos.

Obstacle Detection of Autonomous Land Vehicles for Geological Prospecting

The paper mainly introduces the method of obstacle detection of autonomous land vehicle for geological prospecting. According to the categories and characteristic of obstacles, choosing the applicable sensor is the fundamental of accurately detecting obstacles. Data processing, including information collection, information processing and collecting characteristic parameter obstacle, is the difficulty for accurately detecting obstacles.

Road Edge Detection and Filtering Based on Unclosed Snakes and 2D LIDAR Data

A filtering model of sequential road edges is proposed.The solution of unclosed snakes is discussed based on traditional active contour models,the snakes.Then,it is applied to the smoothing and filtering of road edges detected from 2D LIDAR(LIght Detection And Ranging) data.Several models to depict roads and obstacles are discussed in detail.Based on these models,road edges can be acquired from a single scan 2D LIDAR data quickly and efficiently.Some problems are investigated,such as building and initialization of unclosed snakes model in the road edge filtering scenario,and acquisition of image forces of snakes from LIDAR data.The proposed algorithm is implemented on our ALV(autonomous land vehicle),and the navigation experiments demonstrate that our approach can efficiently reduce the mistakes and errors as well as improve the continuity of road edges.The resulted road edges thus are more faithful to the truth.

Adaptive speed tracking control for autonomous land vehicles in all‐terrain navigation: An experimental study

This paper develops a nonparametric controller with an internal model control (IMC) structure for the longitudinal speed tracking control of autonomous land vehicles by designing a proportional and internal model control (IMC) cascade (P‐IMC) controller. An IMC architecture is employed in the inner control loop by establishing a nonparametric longitudinal dynamical model, whereas a P controller is designed for the outer control loop. An approach for estimating the terrain effects and compensating for the model errors is also introduced. The differences from other nonparametric controllers are discussed, and the stability of the P‐IMC controller is analyzed and validated experimentally. The P‐IMC controller is compared with the SpAM+PI to illustrate its advantages. The experimental results of autonomous all‐terrain driving show the effectiveness of the P‐IMC controller.

Autonomous Land Vehicle Localization Based on Video Compass

The autonomous land vehicle localization is crucial for navigation, local path planning and position control in urban environments. The Video Compass method is proposed for autonomous land vehicle in this paper. First, the lane marking information from monocular camera is extracted. Second, the change of vanishing point is calculated from the lane markings. Third, with the above two conditions the autonomous land vehicle absolute direction could be deduced. This is the technique of the Video Compass proposed in this paper. Finally, the localization is calculated by the Video Compass and the wheel encoder. The algorithm presented in this paper is proved to be both robust and fast by experiments.