|
| A | B | C | D | E | F | G |
|
| [85] | Genetic method | GA-DDS-based optimization model considering internal and external factors | Steering angle, kinematic parameters, dynamic parameters | Sim/video | The proposed model performed 22% and 40% faster than MLP and RF models, respectively | The proposed model has not been evaluated in real-time controlled or dynamic road scenarios |
| [88] | GA-FOPID-based optimization model | PID parameters, integral and derivative order | Simulation | The proposed model achieved an error rate of 1.8 rad | Complex GA structure has been utilized in the proposed model which leads to overshooting and performs worse in undesired scenarios |
| [90] | H∞ method | Presented H∞-based fault lenient lateral controller for the four-wheel steering AVs to improve the effectiveness | Steering angle, cornering stiffness, slip angles, kinematic/dynamic parameters | Achieved lateral error and angular error 0.08 m and 0.14 m, respectively | Simulations carried out with low constant vehicle speed |
| [92] | Proposed H∞ (DYC) inspired system-based steering control design for AVs to improve the efficiency in autonomous path tracking | Vehicle mass, steering angle, cornering stiffness, slip angles, dynamic and kinematic parameters | Decreased the side slip angle and path tracking error effectively | Any practical implementation in the real world has not been performed |
|
| [93] | Neural networks | Presented two-neural networks-based controller for the self-driving | Lane marking, RPM, steering angle | TORCS | The proposed system achieved satisfactory accuracy in the defined simulated track | The proposed system has been evaluated on the static test track with a certain speed and velocity |
| [94] | Fuzzy logic-deep steering neural network-based steering control system for self-driving car | Lane markings, steering angle, and velocity | MATLAB | The proposed system achieved 7.3 MSE in the Udacity challenge environment | The proposed algorithm has not been tested in the dense traffic scenarios |
|
