US Department of Transportation (DOT) operators commonly use adjustable surveillance cameras for traffic monitoring and desire to have an automated traffic counting system by lane. To fill this need, this paper describes an automatic, novel, multiple-ROI (Regions of Interest) lane learning (MRLL) system. It detects lane centers, boundaries, and traffic directions, irrespective of zoom or direction. It finds optimal ROIs without user input by analyzing confidence scores from a chosen Machine Learning (ML) object detector. A simple but effective Continual Learning strategy is used to control the MRLL’s start and stop that optimizes lane counting performance in various real-world conditions: nighttime, extremely harsh weather, or low traffic flow conditions. Tested on 45 varied videos, it achieves an F1_score above 0.79 for lane center detection, 0.88 for lane boundaries, and 94% accuracy in traffic direction detection. This innovative system, which does not rely on lane markings and adapts to camera views, is currently used by the Indiana Department of Transportation for vehicle counting and flow rate estimation in real-world ITS scenarios. Code is available at https://github.com/qiumei1101/Multiple_ROI_lane_learning_system_for_Highway.git .
The problem of the existence and structure of minimum-time control for multiple robot arms cooperatively handling a common object is addressed. The dynamical system is modeled by considering the arms as closed kinematic chains. The conditions of the existence of the time-optimal solution are verified by applying a Lyapunov approach and the concept of the feedback equivalence to linear systems. Furthermore, it is shown that the structure of the minimum-time control law for individual joint actuators requires that at least one of the actuators is always saturated on any finite time subinterval, while the rest of them are either saturated or singular, depending upon the motion configurations and robot parameters. Thus, it is suggested that totally-singular optimal control does not exist in this problem, but that partially-singular control may occur. This theoretical result should be able to provide insight into the dynamic characteristics of the multiple robot system, which can be valuable in the path planning and design specifications of the system.
Vehicle collision avoidance system (CAS) is a control system that can guide the vehicle into a collision-free safe region in the presence of other objects on road. Common CAS functions, such as forward-collision warning and automatic emergency braking, have recently been developed and equipped on production vehicles. However, these CASs focus on mitigating or avoiding potential crashes with the preceding cars and objects. They are not effective for crash scenarios with vehicles from the rear-end or lateral directions. This paper proposes a novel collision avoidance system that will provide the vehicle with all-around (360-degree) collision avoidance capability. A risk evaluation model is developed to calculate potential risk levels by considering surrounding vehicles (according to their relative positions, velocities, and accelerations) and using a predictive occupancy map (POM). By using the POM, the safest path with the minimum risk values is chosen from 12 acceleration-based trajectory directions. The global optimal trajectory is then planned using the optimal rapidly exploring random tree (RRT*) algorithm. The planned vehicle motion profile is generated as the reference for future control. Simulation results show that the developed POM-based CAS demonstrates effective operations to mitigate the potential crashes in both lateral and rear-end crash scenarios.
1 December 1993 in International Journal of Control
A class of optimal control problems in which the control signals are bounded and appear linearly in the hamiltonian function occur in many applications. In the solutions to these problems, the fact that optimal control trajectories may contain discontinuities and possibly singular arcs/sub-arcs makes the problems extremely difficult to solve, if not impossible. This paper presents a novel and efficient computational method for solving such problems. The method consists of two steps. In the first step, the original optimal control problem with possible discontinuities and singular arcs in control is converted into one with continuous and non-singular control trajectories by adding to the performance indexes a perturbed (or weighted) energy term. The resultant boundary value problem can easily be solved for an appropriately large value of the perturbation parameter. In the second step, a continuation method (imbedding method, or homotopy method) is developed to obtain the solution to the original problem by solving a set of initial value problems sequentially and/or in parallel as the perturbation parameter goes to zero. The proposed two-step method is computationally efficient since the resulting two-point boundary value problem is solved only once for a large perturbation parameter and the remaining problem becomes that of solving a set of initial value subproblems. The proposed algorithm is applicable to a large class of optimal control problems with various boundary conditions (e.g. fixed and free terminal time). The practicability of the method is demonstrated by computer simulations on several simple examples, including one with a singular arc along the optimal control trajectory.