A Review of Motion Planning for Highway Autonomous Driving

In recent years, technological advancements have made a promising impact on the development of autonomous vehicles. The evolution of electric vehicles, development of state-of-the-art sensors, and advances in artificial intelligence have provided necessary tools for the academia and industry to develop the prototypes of autonomous vehicles that enhance the road safety and traffic efficiency. The increase in the deployment of sensors for the autonomous vehicle, make it less cost-effective to be utilized by the consumer. This work focuses on the development of full-stack autonomous vehicle using the limited amount of sensors suite. The architecture aspect of the autonomous vehicle is categorized into four layers that include sensor layer, perception layer, planning layer and control layer. In the sensor layer, the integration of exteroceptive and proprioceptive sensors on the autonomous vehicle are presented. The perception of the environment in term localization and detection using exteroceptive sensors are included in the perception layer. In the planning layer, algorithms for mission and motion planning are illustrated by incorporating the route information, velocity replanning and obstacle avoidance. The control layer constitutes lateral and longitudinal control for the autonomous vehicle. For the verification of the proposed system, the autonomous vehicle is tested in an unconstrained environment. The experimentation results show the efficacy of each module, including localization, object detection, mission and motion planning, obstacle avoidance, velocity replanning, lateral and longitudinal control. Further, in order to demonstrate the experimental validation and the application aspect of the autonomous vehicle, the proposed system is tested as an autonomous taxi service.

This paper describes design, vehicle implementation and validation of a motion planning and control algorithm of autonomous driving vehicle for lane change. Autonomous lane change is necessary for high-level autonomous driving. A vehicle equipped with diverse devices like sensors and computer is introduced for implementation and validation of autonomous driving. The autonomous driving system consists of three parts: perception, motion planning and control. In a perception part, surrounding vehicles' states and lane information are estimated. In motion planning part, using these information and chassis information, probabilistic prediction is conducted for ego vehicle and surrounding vehicle separately. And then, driving mode are decided among three modes: lane keeping, lane change and traffic pressure. Driving mode is determined based on a safety distance by predicting states of surrounding vehicles and ego vehicle. If the ego vehicle cannot perform lane change when the lane change is required, the most proper space is selected considering the probabilistic prediction information and the safety distance. Target states are defined based on driving mode and information of surrounding vehicles behaviors. In control part, the distributed control architecture for real time implementation to the vehicle. A linear quadratic regulator (LQR) optimal control and a model predictive control (MPC) are used to obtain the longitudinal acceleration and the desired steering angle. The proposed automated driving algorithm has been evaluated via vehicle test, which has used one autonomous vehicle and two normal vehicles.

A COMMON task for autonomous vehicles is motion planning. Discipline-based design of motion planning algorithms have led to the development and evolution of different techniques to solve specific problems. For instance, the artificial–potential–function technique [1] is a popular method for the motion planning of unmanned ground vehicles (UGV) and robotic manipulators [1–3]. Although this technique has been used for over 30 years, it suffers from the possibility of the vehicle not achieving its goal and difficulties in accommodating various environmental constraints [4]. To overcome such issues, recent developments in motion planning algorithms place heavy emphasis on so-called sampling-based planning techniques such as probabilistic road maps, rapidly exploring random trees, and expansive space trees to name a few [5– 9]. These methods use probabilistic means of connecting the initial configuration to thefinal configuration thereby enabling an improved capacity to achieve the goal and a capability to generate initial feasible paths. In all these techniques, the initial feasible paths do not automatically incorporate vehicle dynamics; hence, path-following control techniques are needed to satisfy the physics of themotion [9]. This is one reason why nonholonomic constraints play such a crucial role in constrained control techniques that are designed to serve pathfollowing systems. There is no doubt that optimal control theory is the most natural framework for solving motion planning problems; however, solving optimal control problems has historically been considered difficult due to the twin curses of dimensionality and complexity. These difficulties are exacerbated in the presence of state constraints; hence, an obstacle-cluttered environment becomes a substantially more difficult problem under the framework of optimal control theory [10,11]. Nonetheless, it is possible to solve simplified motion planning problems wherein the cost function is quadratic or the environment is obstacle free. Because the motion planning techniques developed in robotics applications are unsuitable for flight vehicles, such as launch and reentry, for example, aerospace problems have motivated the development of efficient optimal control algorithms. In aerospace applications, satisfaction of the dynamical constraints is exceedingly important; hence, trajectory and control become fundamentally intertwined. On the other hand, aerospace problems do not have the vast number of path constraints that are common in an obstacle-cluttered environment. In recent years, paradigm-changing advancements have taken place in computational optimal control that challenge conventional wisdom. For instance, onboard the International Space Station, Bedrossian et al. [12] discovered and implemented a revolutionary momentum-dumping approach that they call a zero-propellant maneuver. This discovery was made possible by an application of pseudospectral methods to solve a real-life challenging optimal control problem [13]. Other applications of such advancements are discussed in [14,15] and include the ground test of a revolutionary attitude control concept for the NPSAT1 spacecraft. Motivated by these advancements, we apply pseudospectral methods to develop motion planning algorithms for autonomous vehicles characterized by nonlinear dynamical constraints, an obstacle-cluttered environment, and a need to generate solutions in real time. We consider the problem of generating optimal trajectories for generic autonomous vehicles. Shapes of arbitrary number, size, and configuration are modeled in the form of path constraints in the resulting optimal control problem. The method is tested under various obstacle environments on different platforms such as sea surface vehicles, ground vehicles, and aerial vehicles. The optimality of the computed trajectories is verified by way of the necessary conditions. We show that it is possible to do motion planning for different problems under the unified framework of optimal control and pseudospectral methods [16–19].

This paper describes the design, implementation, and evaluation of a virtual target-based overtaking decision, motion planning, and control algorithm for autonomous vehicles. Both driver acceptance and safety, when surrounded by other vehicles, must be considered during autonomous overtaking. These are considered through safe distance based on human driving behavior. Since all vehicles cannot be equipped with a vehicle to vehicle communications at present, autonomous vehicles should perceive the surrounding environment based on local sensors. In this paper, virtual targets are devised to cope with the limitation of cognitive range. A probabilistic prediction is adopted to enhance safety, given the potential behavior of surrounding vehicles. Then, decision-making and motion planning has been designed based on the probabilistic prediction-based safe distance, which could achieve safety performance without a heavy computational burden. The algorithm has considered the decision rules that drivers use when overtaking. For this purpose, concepts of target space, demand, and possibility for lane change are devised. In this paper, three driving modes are developed for active overtaking. The desired driving mode is decided for safe and efficient overtaking. To obtain desired states and constraints, intuitive motion planning is conducted. A stochastic model predictive control has been adopted to determine vehicle control inputs. The proposed autonomous overtaking algorithm has been evaluated through simulation, which reveals the effectiveness of virtual targets. Also, the proposed algorithm has been successfully implemented on an autonomous vehicle and evaluated via real-world driving tests. Safe and comfortable overtaking driving has been demonstrated using a test vehicle.

The research in this report incorporates the improvement in the autonomous driving capability of self-driving cars in a dynamic environment. Global and local path planning are implemented using the D* path planning algorithm with a combined Cubic B-Spline trajectory generator, which generates an optimal obstacle free trajectory for the vehicle to follow and avoid collision. Model Predictive Control (MPC) is used for the longitudinal and the lateral control of the vehicle. The presented motion planning and control algorithm is tested using Model-In-the-Loop (MIL) method with the help of MATLAB® Driving Scenario Designer and Unreal Engine® Simulator by Epic Games®. Different traffic scenarios are built, and a camera sensor is configured to simulate the sensory data and feed it to the controller for further processing and vehicle motion planning. Simulation results of vehicle motion control with global and local path planning for dynamic obstacle avoidance are presented. The simulation results show that an autonomous vehicle follows a commanded velocity when the relative distance between the ego vehicle and an obstacle is greater than a calculated safe distance. When the relative distance is close to the safe distance, the ego vehicle maintains the headway. When an obstacle is detected by the ego vehicle and the ego vehicle wants to pass the obstacle, the ego vehicle performs obstacle avoidance maneuver by tracking desired lateral positions.

Analysis of pre-crash scenarios and contributing factors for autonomous vehicle crashes at intersections


 
 
 Autonomous Vehicle (AV), also known as self-driving or driverless vehicles, is a technology that intends to replace driving tasks traditionally performed by human drivers. This study aims to gain a better understanding of opinions, concerns, and general acceptance by Malaysian drivers regarding autonomous and self-driving vehicles. The survey questionnaires were adopted from Schoettle & Sivak (2014b) on the topic of public acceptance of driverless vehicles and what would the respondents do when they are not driving. The survey was conducted among the general public in Malaysia inclusive of 19 questions to address topics on public awareness and understanding about AV, benefits of AV, concerns on AV, attitudes toward owning or sharing AV, willingness to pay for AV and their use of free time while riding AV. A total of 520 respondents involved in this survey, however the analysis only considered the respondents who have heard of AV, which amounted to 361 via face to face interview and online survey form. Based on the survey,63.7% were male and 36.3% female. Only 15.2% of the respondents said that they would not ride in a completely self-driving vehicles. AV is also expected to provide a wide range of benefits that attracted most of the respondents. The majority of the respondents expect lower vehicle emission and better fuel economy. Most respondents were also concerned with issues related to AV especially on safety consequences of equipment failure. The survey also shows that most Malaysians are not willing to invest in owning a driverless vehicle. Future studies should continuously consider people’s perception from different perspectives, by taking into account various user viewpoints and characteristics.
 
 

Two Approaches for Path Planning of Unmanned Aerial Vehicles with Avoidance Zones

Motion planning and control of connected and autonomous vehicles (CAVs) for improving traffic efficiency and safety in intersections still meets many challenges due to its dynamic and complex nature. In this article, an innovative collision-free and time-optimal multivehicle motion planning method for the CAVs at unsignalized intersection scenarios is proposed. We systematically analyze the regularity of intersection crossing mode and summarize the overall conflict scenario. To eliminate the vehicles potential collision, a learning-based iterative optimization (LBIO) algorithm is designed to solve the collision-free trajectories generating problem iteratively and offline. The terminal constraint set, terminal cost, and global safe constraints of the LBIO are constructed and updated from the historical data in previous iterations. The algorithm can finally converge to time-optimal trajectories for multivehicle only after several iterations. To apply the trained trajectories into the continuous intersection traffic flow, an online cluster-based motion planning (CBMP) algorithm is developed to coordinate the vehicle velocities and movements in the cooperative control area surrounding the intersection. With an LTV-MPC algorithm for the low-level control, the proposed approach is validated on the SUMO in typical intersection scenarios. The results show that the proposed method allows the potentially conflicting vehicles passing the intersection simultaneously and quickly without waiting, and significantly improves the overall traffic efficiency.

: The main charter of this contract is the implementation and experimentation with motion planning algorithms that emphasize the exact combinatorial and purely geometric approach. Motion planning is considered to be one of the major research areas in robotics, and is one of the main stages in the design and implementation of autonomous intelligent systems, which is an important long-range goal in robotics research. Motion planning is one of the basic capabilities that such a system must possess. In purely geometric terms, the simplest version of the problem can be stated as follows. The system is given complete information about the geometry of the environment in which it is to operate (and of its own structure), and has to process it so that, when commanded to move from its current position to some target position, it can determine whether it can do so without colliding with any of the obstacles around it, and if so plan (and execute) such a motion. These are many variants of the problem. A few of those are: motion planning in environments that are only partially known to the system, compliant motion planning that allows contact with obstacles, which might be unavoidable due to measurement errors, optimal motion planning, motion planning with kino-dynamic constraints, and motion planning amidst moving obstacles. Still, even the simplest, static, and purely geometric version stated above is far from being simple, and poses serious challenges in the design of efficient and robust algorithms.

It is expected that a mixture of autonomous and manual vehicles will persist as a part of the intelligent transportation system (ITS) for many decades. Thus, addressing the safety issues arising from this mix of autonomous and manual vehicles before autonomous vehicles are entirely popularized is crucial. As the ITS system has increased in complexity, autonomous vehicles exhibit problems such as a low intention recognition rate and poor real-time performance when predicting the driving direction; these problems seriously affect the safety and comfort of mixed traffic systems. Therefore, the ability of autonomous vehicles to predict the driving direction in real time according to the surrounding traffic environment must be improved and researchers must work to create a more mature ITS. In this paper, we propose a deep learning-based traffic safety solution for a mixture of autonomous and manual vehicles in a 5G-enabled ITS. In this scheme, a driving trajectory dataset and a natural-driving dataset are employed as the network inputs to long-term memory networks in the 5G-enabled ITS: the probability matrix of each intention is calculated by the softmax function. Then, the final intention probability is obtained by fusing the mean rule in the decision layer. Experimental results show that the proposed scheme achieves intention recognition rates of 91.58% and 90.88% for left and right lane changes, respectively, effectively improving both accuracy and real-time intention recognition and improving the lane change problem in a mixed traffic environment.

Lane change is one of the important operations in motion of an autonomous vehicle. When encountering obstacles or wanting to overtake the vehicle ahead, the autonomous vehicle will make a decision and choose the best path to control the trajectory of motion to perform lane change. In this article, we will present solutions for lane change trajectories, including general path setting, building nonlinear models with states of vehicle speed, acceleration and jerk; building a constraint set to avoid collisions with a minimum safe distance model, which takes into account the potentially collision angle positions during lane change. Simulation results are performed in Matlab simulation environment to demonstrate an effective proposed solution and addressed the disadvantages in the modeling process for lane-changing operations, in order to improve the proactive safety of the motion planning for autonomous vehicles.

Full market penetration for autonomous vehicle requires complete solutions for operation during winter driving conditions. This work addresses three key issues relevant to the dynamic response of an autonomous vehicle when faced with reduced friction due to snow and ice on the road when attempting a double lane change obstacle avoidance maneuver. Two low friction scenarios as well as an improvement to simulation methods are presented. The first low friction scenario an autonomous vehicle may encounter is one in which the road surface friction coefficient is incorrectly assumed to be dry pavement. This scenario could occur in the presence of clear ice on the road which is undetectable by the vehicle until it begins traversing the effected area. In this case, the vehicle must react in a way which maintains vehicle control during the maneuver by adapting to the loss of tractive force at the wheels. This work presents a method for altering the look ahead distance of the common pure pursuit lateral control method for autonomous vehicles. This method stabilizes the vehicle during the maneuvers by dynamically changing the look ahead distance based on cross track error in addition to vehicle velocity. Implementation in the autonomous test vehicle used in this work shows an elimination of off-road occurrences during double lane changes on ice and a 46\% reduction of off-read occurrences during single lane changes. The second low friction scenario an autonomous vehicle may encounter is one in which the road surface friction coefficient is known by the autonomous vehicle through it's own perception or through vehicle to vehicle/infrastructure communication. In this case the vehicle must plan it's path accordingly to ensure the vehicle successfully avoids the obstacle while maintaining control and passenger comfort. This work presents an optimization method which results in a minimum maneuver length across a profile of friction surfaces at a single velocity. This work also investigates the lack of correlation between the autonomous test platform operating on an icy surface and a simulation using a constant coefficient for low friction surfaces. The simulation environment used accurately predicts vehicle dynamic response when simulating operation on dry pavement with a divergence in response on friction values below that of packed snow ($\mu=0.3$). On lower friction surfaces the test vehicle exhibits significant variation in response to steering input. This work presents a stochastic method for representing friction surface in simulation across a grid map to bring simulation

Guest editorial: Decision making and control for connected and automated vehicles

Self-driving vehicles are expected to push towards the evolution of mobility environment in the near future. However, motion sickness has been proven to be one of the main limitations to the successful introduction of fully automated vehicles. Recently, research into motion sickness on autonomous vehicles has just started to get attention. In particular, recent researches have focused on providing solutions to mitigate motion sickness with ideas from the field of human factors and ergonomics, interior design and automotive engineering. However, limited or no representative work has been done in motion planning in terms of motion sickness minimisation. In this respect,this paper presents the application of motion planning in order to minimize motion sickness in self-driving vehicles. By formulating an optimal control problem, the optimum velocity profile is sought for a predefined road path from a specific starting point to a final one. Specific and given boundaries and constrains are applied in order to minimize the motion sickness without compromising the journey time. F or the representation of the motion sickness as a cost function to our optimal control problem, the illness rating is used. Different case studies are investigated by changing the cost functions of our problem as well as varying the lateral manoeuvrability of the vehicle. According to the results, the road with flexible lateral manoeuvrability is found to provide lower sickness as well as shorter journey time when both motion sickness and journey time are consider in the cost function.