`Combined Speed and Steering Control in High Speed Autonomous Ground Vehicles for Obstacle Avoidance Using Model Predictive Control

Abstract:

This paper presents a model predictive control-based obstacle avoidance algorithm for autonomous ground vehicles at high speed in unstructured environments. The novelty of the algorithm is its capability to control the vehicle to avoid obstacles at high speed taking into account dynamical safety constraints through a simultaneous optimization of reference speed and steering angle without a priori knowledge about the environment and without a reference trajectory to follow. Previous work in this specific context optimized only the steering command. In this work, obstacles are detected using a planar light detection and ranging sensor. A multi-phase optimal control problem is then formulated to simultaneously optimize the reference speed and steering commands within the detection range. Vehicle acceleration capability as a function of speed, as well as stability and handling concerns such as preventing wheel lift-off are included as constraints in the optimization problem, whereas the cost function is formulated to navigate the vehicle as quickly as possible with smooth control commands. Simulation results show that the proposed algorithm is capable of safely exploiting the dynamic limits of the vehicle while navigating the vehicle through sensed obstacles of different size and number. It is also shown that the proposed variable speed formulation can significantly improve performance by allowing navigation of obstacle fields that would otherwise not be cleared with steering control alone.

EXISTING  SYSTEM:

Prior work by the authors has started to address these limitations. First, the role of model fidelity in the MPC framework has been evaluated in [19], [20]. Second, a nonlinear MPC-based obstacle avoidance algorithm has been developed for large-size, high-speed AGVs in [21], [22]. The environment is perceived only through a planar LIDAR sensor. The corresponding sensor data processing algorithm has also been developed and described in [23]. The algorithm can achieve an optimal and smooth operation of AGVs at high speed through unstructured environments without collision while ensuring vehicle dynamical safety; i.e., without wheel lift-off. However, the formulation assumes that the vehicle longitudinal speed is maintained constant, which can limit the mobility performance and the obstacle fields that can be cleared with this algorithm.

PROPOSED  SYSTEM:

In this paper, we extend the previous work and develop a novel MPC formulation that simultaneously optimizes both the longitudinal speed and steering control commands for high speed obstacle avoidance taking into account dynamical safety. The novelty of the formulation is threefold: (1) A varying prediction horizon MPC is used to achieve a fixed distance prediction. This is prompted by two features of the proposed system. First, the terminal point of the planned trajectory is constrained at the LIDAR’s maximum detection range in an effort to fully utilize as much information from the LIDAR as possible. Second, the variable vehicle speed necessarily leads to a variable prediction horizon with the previous constraint. (2) The effects of the powertrain and brake systems are taken into account through the relationship between acceleration and speed and the bounds on longitudinal jerk, acceleration, and speed. The vehicle’s acceleration capability varies with the speed resulting from the powertrain and brake systems. To generate a speed profile that can be tracked by the vehicle, the algorithm uses an offline generated look-up table to account for the acceleration and deceleration limitations. (3) The no-wheel-lift-off requirement is considered through both hard and soft constraints using equations with empirical parameters that can predict tire vertical loads. A hard constraint bounds the vertical loads to be greater than a specified minimum threshold. A soft constraint is also used to provide a smooth approach to this threshold to prevent overshoot.

CONCLUSIONS:

This paper considers high-speed AGVs in unstructured environments without a priori information about the obstacles and presents a new MPC-based obstacle avoidance algorithm that optimizes the longitudinal speed and steering angle simultaneously to navigate the AGV safely and as quickly as possible to the target location. To this end, the algorithm is capable of exploiting the dynamic limits of the vehicle to maximize the vehicle’s mobility performance. A multi-stage OCP formulation is used to incorporate the obstacle data obtained from the on-board LIDAR sensor. The time length of the prediction horizon of the MPC is varying because a variable-speed trajectory is planned till the end of the sensor range. The powertrain and brake dynamics are taken into consideration through the bounds on vehicle longitudinal speed, acceleration and jerk. The dynamical safety requirement is accounted for by enforcing a positive vertical load on all four tires as hard constraints. One term in the cost function also aims to provide a smooth approaching to the vertical load threshold. Three sets of numerical simulations are conducted to demonstrate the effectiveness of the algorithm. The conclusion is that the newly developed algorithm with variable speed and steering commands not only improves the performance of the vehicle by allowing it to operate closer to its dynamical limits, but also enables the safe clearance of obstacle fields that may not be cleared with steering control alone

REFERENCES:

[1] S. M. LaValle, Planning algorithms, Cambridge University Press, 2006.
[2] C. E. Beal and J. C. Gerdes,“Model predictive control for vehicle stabilization at the limits of handling”, IEEE Transactions on Control Systems Technology, vol. 21, pp. 1258-1269, 2013.

[3] B. D. Luders, S. Karaman, and J. P. How, “Robust sampling-based motion planning with asymptotic optimality guarantees,” in Proceedings of the AIAA Guidance, Navigation, and Control Conference (GNC), 2013.

[4] O. Khatib, “Real-time obstacle avoidance for manipulators and mobile robots,” International Journal of Robotics Research, vol. 5, pp. 90-98, 1986.

[5] S. Shimoda, Y. Kuroda, and K. Iagnemma, “High-speed navigation of unmanned ground vehicles on uneven terrain using potential fields,” Robotica, vol. 25, pp. 409-424, 2007.

[6] A. Hussein, H. Mostafa, M. Badrel-Din, O. Sultan, and A. Khamis, “Meta-heuristic optimization approach to mobile robot path planning,” in Proceedings of the International Conference on Engineering and Technology (ICET), 2012.

[7] P. Ogren and N. E. Leonard, “A convergent dynamic window approach to obstacle avoidance,” IEEE Transactions on Robotics, vol. 21, pp. 188-195, 2005.

[8] Y. Gao, T. Lin, F. Borrelli, E. Tseng, and D. Hrovat, “Predictive control of autonomous ground vehicles with obstacle avoidance on slippery roads,” in Proceedings of the Dynamic Systems and Control Conference (DSCC), 2010.

[9] F. Allgwer, and A. Zheng, Nonlinear model predictive control, Springer, 2000. [10] J. M. Park, D. W. Kim, Y. S. Yoon, H. J. Kim, and K. S. Yi, “Obstacle avoidance of autonomous vehicles based on model predictive control,” Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, vol. 223, pp. 1499-1516, 2009.