Making Bertha Drive—An Autonomous Journey on a Historic Route

The Japan Digital Road Map Association (JDRMA) was founded to establish the standards for digital road map (DRM) and to produce a DRM database in Japan. The JDRMA is now in the process of refining the database using 1/25,000 scale topographical maps. The JDRMA supplies users with an updated version of the database every year. Car navigation is currently the largest application of DRM. Police departments, fire departments, ambulance services, and taxi companies use location systems that use DRM. Significant application of DRM is expected to come in the future in the planning and research fields. Road sign checking system, traffic accident analysis system, road census reference system and logistics system are now being studied. As the applications expand, more advanced DRMs must be made available, To meet this requirement, development of more detailed and precise DRMs and linkage to dynamic information systems is being pursued.

This paper describes the design of a fully automated driving algorithm for automated driving in complex urban scenarios and motorways with a satisfactory safety level. The proposed algorithm consists of the following three steps: surround recognition, motion planning, and vehicle control. The surround recognition system consists of three main modules: object classification, vehicle/non-vehicle tracking and dynamic drivable area determination. All system modules utilize information from potentially commercializable sensors such as vision sensors, radars, lidar and vehicle sensors. The objective of the motion planning module is to derive an optimal trajectory as a function of time and the surround recognition results. A dynamic drivable area is represented as a complete driving corridor that leads to the destination while making sure all objects are outside the left or right corridor bounds. In the case of moving objects such as other traffic participants, their behaviors are anticipated within the dynamic drivable area. The optimal trajectory planning uses the dynamic drivable area as a safety constraint and computes a trajectory in which the vehicle stays in its safe bounds considering the driver?s pattern and characteristics based on predicted risk potential. The developed algorithm has been evaluated by computer simulation and vehicle tests on urban roads and motorways.

Accurate state recognition of traffic lights is important for a safe urban drive of autonomous vehicles. A traffic light with lower luminance than its ambient is difficult to recognize particularly in a distant. This paper proposes a strategy that improves accuracy of signal state recognition of distant traffic lights with low luminance. Digital map is used to get candidate regions of traffic lights on a captured image. Each candidate is cropped and resized to equalize the size of lamp area on the image. On the resized images, brightness channel image is modified by enhancing pixels that have hue value in the range of traffic signal color. Then, high value pixels on the SV m image, which is the muitilpied image of the satulation channel and the modified brightness channel, is regarded as candidates of lamp area. This strategy attempts to emphasize traffic signal lights darker than its ambient. Experiments in real driving data showed the performance of the proposed method.

Lane-level localization is critical for autonomous vehicles (AVs). However, complex urban scenarios, particularly tunnels, pose significant challenges to AVs’ localization systems. In this paper, we propose a fusion localization method that integrates multiple mass-production sensors, including Global Navigation Satellite Systems (GNSSs), Inertial Measurement Units (IMUs), cameras, and high-definition (HD) maps. Firstly, we use a novel electronic horizon module to assess GNSS integrity and concurrently load the HD map data surrounding the AVs. This map data are then transformed into a visual space to match the corresponding lane lines captured by the on-board camera using an improved BiSeNet. Consequently, the matched HD map data are used to correct our localization algorithm, which is driven by an extended Kalman filter that integrates multiple sources of information, encompassing a GNSS, IMU, speedometer, camera, and HD maps. Our system is designed with redundancy to handle challenging city tunnel scenarios. To evaluate the proposed system, real-world experiments were conducted on a 36-kilometer city route that includes nine consecutive tunnels, totaling near 13 km and accounting for 35% of the entire route. The experimental results reveal that 99% of lateral localization errors are less than 0.29 m, and 90% of longitudinal localization errors are less than 3.25 m, ensuring reliable lane-level localization for AVs in challenging urban tunnel scenarios.

Japan Digital Road Map Association (JDRMA) started to construct the digital road map database for the whole Japan in 1988. The first database was completed in 1989, and consequently the updating started in 1990. There are two kinds of data sources for updating the JDRMA database. One is the newly issued topographic maps, which include both roads and other features but are not updated every year. The other is the information from the road administrators; it is used to update the road networks. Every year the information of the newly opened or improved roads is collected and used to update the road network in the database. Other information is also used to supplement the two main information sources. Documents of transportation facilities other than roads and information from the users are utilized. The investigation to improve the updating procedure is continued. >

<div class="section abstract"><div class="htmlview paragraph">Autonomous vehicle is a vehicle capable of sensing its environment and taking decisions automatically with no human interventions. To achieve this goal, ADAS (Advance Driving Assistance System) technologies play an important role and the technologies are improving and emerging. The sensing of environment can be achieved with the help of sensors like Radar and Camera. Radar sensors are used in detecting the range, speed and directions of multiple targets using complex signal processing algorithms. Radar with long range and short range are widely used in the autonomous vehicles. Radar sensors with long range can be used to realize features like Adaptive Cruise Control, Advance Emergency Brake Assist. The short-range radar sensors are used for Blind Spot Monitoring, Lane Change Assist, Rear/Front Cross Traffic Alert and Occupant Safe Exit. To realize the Autonomous vehicle functionalities four short range radar sensors are required, two on front and two on rear (left and right).</div><div class="htmlview paragraph">This paper presents a detailed study on different solutions and methodologies for simulating radar signals and validation of short-range radar-based features. The various methodologies involve simulating of radar signals using Radar Target Simulator, Bypassing the radar sensing and injecting the radar raw signals, Bypassing the radar sensing injecting the target objects and Field and track testing in the real time environment. This paper also describes the challenges and comparison of different solutions.</div></div>

The autonomous vehicle is an automated system equipped with features like environment perception, decision-making, motion planning, and control and execution technology. Navigating in an unstructured and complex environment is a huge challenge for autonomous vehicles, due to the irregular shape of road, the requirement of real-time planning, and the nonholonomic constraints of vehicle. This paper presents a motion planning method, based on the Radial Basis Function (RBF) neural network, to guide the autonomous vehicle in unstructured environments. The proposed algorithm extracts the drivable region from the perception grid map based on the global path, which is available in the road network. The sample points are randomly selected in the drivable region, and a gradient descent method is used to train the RBF network. The parameters of the motion-planning algorithm are verified through the simulation and experiment. It is observed that the proposed approach produces a flexible, smooth, and safe path that can fit any road shape. The method is implemented on autonomous vehicle and verified against many outdoor scenes; furthermore, a comparison of proposed method with the existing well-known Rapidly-exploring Random Tree (RRT) method is presented. The experimental results show that the proposed method is highly effective in planning the vehicle path and offers better motion quality.

Use of autonomous vehicles aims to eventually reduce the number of motor vehicle fatalities caused by humans. Deep learning plays an important role in making this possible because it can leverage the huge amount of training data that comes from autonomous car sensors. Automatic recognition of traffic light and vehicle signal is a perception module critical to autonomous vehicles because a deadly car accident could happen if a vehicle fails to follow traffic lights or vehicle signals. A practical Traffic Light Recognition (TLR) or Vehicle Signal Recognition (VSR) faces some challenges, including varying illumination conditions, false positives and long computation time. In this chapter, we propose a novel approach to recognize Traffic Light (TL) and Vehicle Signal (VS) with high dynamic range imaging and deep learning in real-time. Different from existing approaches which use only bright images, we use both high exposure/bright and low exposure/dark images provided by a high dynamic range camera. TL candidates can be detected robustly from low exposure/dark frames because they have a clean dark background. The TL candidates on the consecutive high exposure/bright frames are then classified accurately using a convolutional neural network. The dual-channel mechanism can achieve promising results because it uses undistorted color and shape information of low exposure/dark frames as well as rich texture of high exposure/bright frames. Furthermore, the TLR performance is boosted by incorporating a temporal trajectory tracking method. To speed up the process, a region of interest is generated to reduce the search regions for the TL candidates. The experimental results on a large dual-channel database have shown that our dual-channel approach outperforms the state of the art which uses only bright images. Encouraged by the promising performance of the TLR, we extend the dual-channel approach to vehicle signal recognition. The algorithm reported in this chapter has been integrated into our autonomous vehicle via Data Distribute Service (DDS) and works robustly in real roads.

<div class="section abstract"><div class="htmlview paragraph">Autonomous vehicles must guarantee safety in all road conditions, including driving on wet roads. Aquaplaning (or hydroplaning) is a phenomenon known since the beginning of automotive history, never solved by an active safety system. Currently, no countermeasure system on the market is able to effectively counteract aquaplaning: ABS, ESP or TCS are still inefficient in overcoming this situation. Latest statistical data confirm that the higher percentage of accidents, injuries and deaths are caused by wet road conditions. The aquaplaning happens when the water on the road is too much and the tires start to float causing the instantaneous loss of control. Such phenomenon occurs in human-driven vehicles, with the responsibility of the driver, but in autonomous vehicles (e.g. Level 5), the responsibility for the safety depends on the car and the reduction of the speed is not a solution. In this work, it is presented the second version of our anti-aquaplaning active safety system, which can identify, without any new sensor, if the vehicle is subject to aquaplaning. When the software detects the phenomenon, the hydraulic system, designed to have a very short reaction time, makes an intelligent injection of water in front of the tires restoring the adherence of any type of vehicle (human-driven or autonomous) in a few milliseconds. It will also be shown how the system could be integrated with sensorized tires (tested for the first time in the automotive history). Moreover, the effects coming from the synergy between the system, ADAS and 5G technology will finally be highlighted in terms of improvements for the future smart mobility.</div></div>

According to the world health organization, there were 1.25 million road accident deaths globally in 2013. To decrease this count, it has been said that autonomous vehicles can play a major role. The autonomous vehicle uses different sensors such as LIDAR, RADAR, Ultrasonic sensors, and cameras for detection of lane lines, objects, traffic sign, traffic lights, and driving path. There are several successful projects on autonomous vehicles, but the main limitations are high costs and regulations. The camera can be a cost-effective solution as compare to LIDAR in all the basic detection needed for autonomous driving. The proper placement of sensors on the vehicle plays an important role in efficient design, working and reducing the cost of the overall autonomous vehicle system. This paper discusses the parameters considered while deciding the positioning of the camera, and the geometric calculation for the best positioning of a monocular camera on an autonomous vehicle.

Motion planning remains a fundamental problem in robotics. Sampling-based algorithms use randomization to allow efficient solutions to this complex problem. As mobile robots and autonomous vehicles become more prevalent in everyday life, motion planning must be applied to increasingly challenging scenarios. Safety has become a paramount concern in motion planning for ensuring robotic applications enrich human lives. To date, many motion planning techniques to increase safety in the face of uncertain and dynamic environments have been developed. This dissertation first addresses distributional safety of Rapidly-Exploring Random Trees (RRT) through our algorithm W-Safe RRT. To acknowledge distributional uncertainty and poor modeling, W-Safe RRT uses the Wasserstein metric to provide a probabilistic bound on the distributional distance between a robot and obstacles. Human-interpretable environmental agent classification allows online safety margin adaptation. We propose and analyze an integrating region method for online classification that increases actor labeling accuracy based on behavioral feature values when compared to state of the art methods. The method performs class assignments based on local maximum likelihood in a created behavioral feature-space, allowing a notion of classification uncertainty. Model-based methods with safety guarantees can quickly become computationally in tractable, especially with multiple agents, higher dimensions, and plentiful unknowns. Sampling based algorithms have been parallelized for computation with multi-core computers and GPUs. We consider the use of quantum algorithms and computers for sampling-based motion planning for the first time. Quantum computing performs operations on superpositions of states and can solve certain problems much more efficiently than classical computers, but introduces previously unseen challenges. With Quantum-RRT, we recast the motion planning problem into a database-search structure and use Quantum Amplitude Amplification to find reachable states in the database with a quadratic performance increase over classical methods. We address two error sources with this method: quantum measurement and quantum oracle errors. We then extend this method to Parallel Quantum-RRT, which uses a manager-worker architecture with multiple parallel quantum workers to increase database search efficiency. We compare algorithm architectures and characterize probabilities of multiple workers finding solutions. Lastly, we test in simulation the quantum algorithms against classical versions in a wide variety of scenarios, concluding that a similar parallelization improvement is to be found in the quantum case as was found in the parallelization of classical RRT.

Traffic lights detection and their state recognition is a crucial task that autonomous vehicles must reliably fulfill. Despite scientific endeavors, it still is an open problem due to the variations of traffic lights and their perception in image form. Unlike previous studies, this paper investigates the use of inaccurate and publicly available GIS databases such as OpenStreetMap. In addition, we are the first to exploit conic section geometry to improve the shape cue of the traffic lights in images. Conic section also enables us to estimate the pose of the traffic lights with respect to the camera. Our approach can detect multiple traffic lights in the scene, it also is able to detect the traffic lights in the absence of prior knowledge, and detect the traffics lights as far as 70 meters. The proposed approach has been evaluated for different scenarios and the results show that the use of stereo cameras significantly improves the accuracy of the traffic lights detection and pose estimation.

The paper focuses on the development of GDF as a standard for digital road maps. GDF is a geographical data format in which the data are defined in three different levels: level O contains the geometry and the topology of the information; level 1 defines real world objects; and on level 2, functionally related level 1 features are grouped into higher order objects. The creation of the European digital road map, which was made possible by the standard and on the expected growth of mass market applications such as car information systems, is also discussed.

The GPS-based navigation systems provide an accurate positioning of land-vehicles but this service is not always continuous in urban areas. These systems are usually enhanced with dead reckoning sensors, like odometers, and the use of a digital road map can improve the vehicle location when GPS fails. In this work, we focus on these GPS outages where the main problem is to combine another available measurements intelligently. We present here a global estimation method to solve the fusion problem of the GPS, odometer and digital road map measurements. It relies on an unscented Kalman filter whose main benefit is to deal with non-linear equations directly. In the case of an urban transport scenario, the comparison with an extended Kalman filter shows that the vehicle positioning is improved without requiring more computing time.

Location based services (LBS) are capturing global spotlights increasingly, in which users tend to pick up navigation service as their favorites. This paper examines the dynamic routing problems based on real-time traffic information in LBS. At first, the overall structure of dynamic navigation system is illustrated, which is composed of four ingredients: traffic data collection, digital road map, static routing and dynamic re-routing. Traffic data collection is the task of traffic information center (TIC), where various sources of data are fused into link travel times that are used as criteria to choose candidates in path determining. Digital road map includes digital road network and historical traffic data, where the topology relationship of road network is elementarily explored and the method to represent statistical traffic data is referred to. Routing with historical traffic data and dynamic re-routing with live traffic data are two main functions of dynamic navigation service. The flow of dynamic re-routing algorithm is exhibited. Lastly, a first-in-first-out (FIFO) time-depend shortest path algorithm is discussed for dynamic routing, which adopts the adaptive A^* algorithm based on binary heap priority queue and RB-tree.