ACC

Overview#

The integrated ACC car-following model is based on the work of Milanés & Shladover [1] and Xiao, Wang & van Arem [2], wherein the developed control law in the ACC control algorithm is explicitly divided into three modes: (i) speed (or cruising) control, (ii) gap control and (iii) gap-closing control. A fourth mode (i.e. collision avoidance mode) has been introduced within the project TransAID.

Note, that the additional modes (iii) and (iv) aim to enable stable and safe car-following behaviour in SUMO, whereas the original publication [1] reveals string instablities in experimental data and simulations for vehicle platoons with ACC controllers (see String Instabilities in ACC Platoons).

see also CACC model.

(i) Speed control mode#

The speed control mode is designed to maintain the pre-defined by the driver desired speed and is activated when there are no preceding vehicles in the range covered by the sensors or preceding vehicles exist in a spacing larger of 120 m.

(ii) Gap control mode#

The gap control mode aims to maintain a constant time gap between the ACC-equipped vehicle and its predecessor and is activated when the gap and speed deviations (with respect to the preceding vehicle) are concurrently smaller than 0.2 m and 0.1 m/s, respectively.

(iii) Gap-closing control mode#

The gap-closing controller enables the smooth transition from speed control mode to gap control mode and is triggered when the spacing to the preceding vehicle is smaller than 100 m. If the spacing is between 100 m and 120 m, the ACC-equipped vehicle retains the previous control strategy to provide hysteresis in the control loop and perform a smooth transfer between the two strategies.

(iv) Collision avoidance control mode#

The collision avoidance mode prevents rear-end collisions when safety critical conditions prevail. This mode is activated when the spacing to the preceding vehicle is smaller than 100 m and the gap deviation is negative.

Also, if the followSpeed computed by the ACC model grows higher than the safe followSpeed as computed by the default Krauss model by a given margin (configured by collisionAvoidanceOverride), the speed is limited to the value of Krauss-speed + margin. The override margin defaults to 2m/s.

String Instabilities in ACC Platoons#

Milanés & Shladover [1] showed in their experiments that the proposed ACC controller is string instable. Although ACC controllers can be calibrated to be string stable, e.g. Gunter et al. [3] raise concern that commercial vehicles mostly deploy ACC controllers that are string instable due to driving comfort reasons.

To investigate stablity effects of vehicle platoons deploying the original ACC model with SUMO, it's possible to disable the modes (iii) and (iv) by resetting the following parameters to:

Mode Attribute Value
iii gapClosingControlGainSpeed 0.07
iii gapClosingControlGainSpace 0.23
iv collisionAvoidanceGainSpeed 0.07
iv collisionAvoidanceGainSpace 0.23

Note, that the parameter collisionAvoidanceOverride (default: 2) affects the mode (iv) by overriding the followSpeed when deemed unsafe by the given margin. To eliminate this case, which slightly affects the platooning behaviour, but not the string instability itself, you this parameter may be set to to a much larger value (i.e. 100) to avoid the effect.

The default parameters of the ACC model can be found here: Car-Following Model Parameters

Given the parametrization from the table, you could replicate the platoon oscilation as in Fig.12 in [1] with SUMO:

ACC platoon

Notes#

  • The implemented model can be found in <SUMO_HOME>/src/microsim/cfmodels/MSCFModel_ACC.cpp.
  • Literature on the developed ACC driving model and its implementation can be found here.
  • This part of SUMO was developed and extended within the project TransAID.
  • The model is primarily intended for use in specific traffic situations.
  • When there is no leader vehicle, the model uses the same speed as the Krauss model to approach junctions and speed limits

Caution

The model is known to produce collisions at the default step-length of 1s. Better results can be achieved by setting a lower step length (e.g. 0.1s).

References#

  1. Milanés, V., & Shladover, S. E. (2014). Modeling cooperative and autonomous adaptive cruise control dynamic responses using experimental data. Transportation Research Part C: Emerging Technologies, 48, pp. 285–300. https://doi.org/10.1016/j.trc.2014.09.001
  2. Xiao, L., Wang, M., & van Arem, B. (2017). Realistic Car-Following Models for Microscopic Simulation of Adaptive and Cooperative Adaptive Cruise Control Vehicles. Transportation Research Record: Journal of the Transportation Research Board, 2623, pp. 1–9. https://doi.org/10.3141/2623-01
  3. G. Gunter et al., "Are Commercially Implemented Adaptive Cruise Control Systems String Stable?," in IEEE Transactions on Intelligent Transportation Systems, vol. 22, no. 11, pp. 6992-7003, Nov. 2021. https://doi.org/10.1109/TITS.2020.3000682