Stochastic Model Predictive Velocity Control for Automobiles Considering Uncertainty of Nearby Vehicles

Stochastic Model Predictive Obstacle Avoidance with Velocity Reduction in Accordance with Chance Constraints in Crowded Environments

“…Note that the subscript v denotes longitudinal motion parameters. We obtain the input sequence 𝑈 v,𝑖 ≔ {𝑢 v,0 , 𝑢 v,1 , … , u v,𝐻 v −1 } that minimizes (17) with an optimization problem. The problem is represented as follows: minimize 𝐽 v , with respect to u v,𝑘 (𝑘 = 0, 1, ⋯ , 𝐻 v − 1), subject to…”

“…This issue has been explicitly addressed in previous research (12,13,14) , which provides a backup or best-effort solution, that does not achieve optimality. Another approach is to reduce the vehicle velocity to decrease the predicted collision damage (15)(16)(17) . In the MPC approach (17) , by dividing the system into longitudinal and lateral dynamics, the relative speed is explicitly evaluated as a stochastic expectation for controlling the velocity.…”

“…Another approach is to reduce the vehicle velocity to decrease the predicted collision damage (15)(16)(17) . In the MPC approach (17) , by dividing the system into longitudinal and lateral dynamics, the relative speed is explicitly evaluated as a stochastic expectation for controlling the velocity. This approach enables us to separate the problem into two subproblems: maintaining sufficient stochastic distance to other traffic participants and controlling the ego-vehicle speed to consider the collision risk.…”

An approach for numerical simulating and processing of measured electrical signals from board sensors installed on wheeled electro-mechanical platforms

“…It is naturally, the automation of wheeled platforms is significantly restricted by opportunities of instrumental measures for estimating their motion state and their interactions with an environment necessary to define the required controls. Really, a lot of the modern researches are about different kinds problems dealt about controlling, measuring and defining the motion state of the wheeled platforms including the general autonomous navigation and motion safety [3,7,8], the path's trajectory [1,9], the current position and the attitude [4,10] as well as by the velocity [8] and the acceleration [11]. Besides, there are a lot of researches dealt with defining the mechanical interactions between the wheels and the soils [12,13] as well as with estimating the aerodynamic [14] and control forces [15].…”

“…Numerical processing of the measured signals is based on using the mathematical models providing the relations between the directly measured signals from the board sensors and the wheeled platforms parameters, so these mathematical models must having all necessary properties agreed with their using purposes, and constructing such mathematical models can be the difficult problem in general. Although, the mathematical models are widely discussed for numeric processing of measuring signals from the board sensors installed on wheeled platforms, but the most researches dealt with the particular kinds of mathematical models, measurements and their processing like in [1,[8][9][10][15][16][17][18][19][20]. At the same time, there are no the conventional generalized approaches of mathematical modelling fit for the numeric processing of the measurement results.…”