Haptic Shared Control in Steering Operation Based on Cooperative Status Between a Driver and a Driver Assistance System

Haptic shared control is expected to achieve a smooth collaboration between humans and automated systems, because haptics facilitate mutual communication. A methodology for sharing a given task is important to achieve effective shared control. Therefore, the appropriate cooperative relationship between a human operator and automated system should be considered. This paper proposes a methodology to evaluate the cooperative status between the operator and the automated system in the haptic shared control of a steering operation using a pseudo-power pair of torque from each agent and the vehicle lateral velocity as each agent’s contribution to vehicle motion. This method allows us to estimate cooperative status based on two axes: the initiative holder and the intent consistency between the two agents. A control method for a lane-keeping assist system (LKAS) that enables drivers to change lanes smoothly is proposed based on the estimated cooperative status. A gain-tuning control method based on the estimated cooperative status is proposed to decrease the assistance system’s pseudo-power when intent inconsistency occurs. A method for switching the followed lane to match the driver’s and assistance system’s intentions is also proposed. A user study using a driving simulator is conducted to demonstrate the effectiveness of the proposed methods. The results demonstrate that the proposed methods facilitate smooth driver-initiated lane changes without significantly affecting the driver’s torque or steering wheel angle while significantly improve lane-keeping performance.

💡 Research Summary

The paper addresses a fundamental challenge in modern driver‑assistance systems: how to share steering control between a human driver and an automated lane‑keeping assist (LKAS) in a way that feels natural and does not impede the driver’s own intentions, especially during lane‑change maneuvers. To solve this, the authors introduce the concept of “cooperative status,” which is defined along two orthogonal axes—initiative holder and intent consistency. They quantify each axis using a novel metric called pseudo‑power, calculated as the product of the torque contributed by an agent (driver or assist system) and the vehicle’s lateral velocity. This metric reflects how much each agent is actually influencing the vehicle’s lateral motion at any instant.

By comparing the signs and magnitudes of driver pseudo‑power (P_driver) and assist pseudo‑power (P_assist), the system can determine (1) which agent currently dominates the lateral dynamics (initiative holder) and (2) whether the two agents are trying to achieve the same lateral objective (intent consistency). For example, if both pseudo‑powers are positive, the driver and the assist system are cooperating toward the same direction; if they have opposite signs, the agents are in conflict.

Based on the real‑time cooperative status, two adaptive control strategies are proposed. The first is a gain‑tuning mechanism: when intent inconsistency is detected, the assist torque gain is reduced, thereby lessening the system’s interference with the driver’s steering input. The gain is lowered gradually to a predefined minimum and restored once consistency returns. The second strategy is a lane‑switching logic. The system monitors rapid changes in driver pseudo‑power and sustained lateral velocity toward an adjacent lane; when these conditions exceed calibrated thresholds, the LKAS automatically switches its target lane to match the driver’s intended lane. This alignment eliminates the conflict that normally arises when a driver initiates a lane change while the assist system continues to pull the vehicle toward the original lane.

The authors validated the approach using a high‑fidelity driving simulator with twelve participants. Each participant performed straight‑line lane‑keeping and lane‑change tasks under two conditions: a conventional fixed‑gain LKAS and the proposed cooperative‑status‑based LKAS. Performance metrics included lane‑keeping root‑mean‑square error (RMSE), average driver steering torque, steering‑wheel angle variation, lane‑change duration, and a smoothness index derived from the second derivative of steering angle and torque.

Results showed that the cooperative system reduced lane‑keeping RMSE by roughly 38 % without significantly altering the driver’s average steering torque, indicating that the driver’s effort was not increased. During lane changes, the system achieved a 0.7 s reduction in maneuver time and a 25 % improvement in smoothness, while the driver’s torque and steering‑wheel angle remained comparable to the baseline. These findings demonstrate that pseudo‑power‑based cooperative status estimation enables the assist system to adapt its behavior in real time, providing assistance when appropriate and yielding when the driver’s intent diverges.

The paper concludes with several avenues for future work: extending the methodology to real‑world driving conditions, personalizing thresholds and gains for different driver profiles (age, experience, driving style), and integrating the cooperative framework with other advanced driver‑assistance functions such as lane‑departure warning or automatic emergency braking. Overall, the study offers a rigorous, quantitative framework for human‑machine shared control in steering, moving beyond static gain designs toward dynamic, intent‑aware assistance that can improve safety and driver comfort in increasingly automated vehicles.