Adaptive controller

1. An autonomous lawn mower comprising:

a planner configured to generate target data indicating a target state for the autonomous lawn mower;

a kinematics component configured to:

receive the target data from the planner;

receive first state data indicating a current state of the autonomous lawn mower, the current state including at least one of a current position, a current orientation, a current speed, or a current rotational rate;

determine a difference between the current state of the autonomous lawn mower and the target state;

determine that the difference is less than a threshold amount of difference; and

based at least in part on the determination that the difference is less than the threshold amount of difference, generate compensation data indicating one or more of a compensation speed or a compensation rotational rate;

a dynamic component configured to:

receive the compensation data from the kinematics component;

receive second state data; and

based at least in part on the compensation data and the second state data, determine a control signal for controlling movement of the autonomous lawn mower to reach the target state; and

a drive unit configured to control movement of the autonomous lawn mower based at least in part on the control signal.

2. The autonomous lawn mower of claim 1 , wherein the autonomous lawn mower further comprises a state estimation component configured to update the first state data based at least in part on the control signal.

3. The autonomous lawn mower of claim 1 , wherein the dynamic component is configured to determine the control signal based at least in part on a difference of the compensation data and the second state data.

4. The autonomous lawn mower of claim 1 , wherein the second state data indicates the current speed and the current rotational rate.

5. The autonomous lawn mower of claim 1 , wherein the dynamic component is configured to determine the control signal by:

determining a derivative of a desired state vector, the desired state vector being based at least in part on the compensation data;

multiplying a state matrix and a dynamic parameter to obtain a first product, the state matrix being based at least in part on the second state data;

multiplying a gain matrix and a first difference of a current state vector and the desired state vector to obtain a second product, the current state vector being based at least in part on the second state data;

subtracting the first product and the second product from the derivative of the desired state vector to obtain a second difference;

multiplying an inverse of a dynamic parameter function and the second difference to obtain a third product; and

determining an amount of torque based at least in part on the third product;

wherein the control signal indicates the amount of torque.

6. The autonomous lawn mower of claim 1 , wherein that the planner is configured to:

determine that the current position of the autonomous lawn mower is more than a threshold distance from the target state; and

generate new target data based at least in part on the determination that the current position of the autonomous lawn mower is greater than or equal to the threshold distance from the target state.

7. The autonomous lawn mower of claim 1 , further comprising:

a dynamic parameter component configured to determine a dynamic parameter related to at least one of a mass associated with the autonomous lawn mower or inertia of the autonomous lawn mower;

wherein the dynamic component is configured to determine the control signal based at least in part on the dynamic parameter.

8. A method comprising:

receiving a target state for an autonomous lawn mower;

determining a first difference between a first current state of the autonomous lawn mower and the target state, the first current state including at least one of a current position, a current orientation, a current speed, or a current rotational rate;

determining that the first difference is less than a threshold amount of difference;

determining, as compensation data and based at least in part on the determination that the first difference is less than the threshold amount of difference, one or more of a compensation speed or compensation rotational rate;

determining a second difference between a second current state and the compensation data;

determining, based at least in part on the second difference, an amount of torque; and

controlling the autonomous lawn mower in accordance with the amount of torque.

9. The method of claim 8 , further comprising:

determining an estimated state of the autonomous lawn mower based at least in part on the amount of torque; and

determining a subsequent amount of torque to apply based at least in part on the estimated state of the autonomous lawn mower.

10. The method of claim 8 , wherein the determining the amount of torque includes:

determining a derivative of a desired state vector;

multiplying a state matrix and a dynamic parameter to obtain a first product, the state matrix being based at least in part on the second current state;

multiplying a gain matrix and a first difference of a current state vector and the desired state vector to obtain a second product, the current state vector being based at least in part on the second current state;

subtracting the first product and the second product from the derivative of the desired state vector to obtain a second difference;

multiplying an inverse of a dynamic parameter function and the second difference obtained from the subtracting to obtain a third product; and

determining the amount of torque based at least in part on the third product.

11. The method of claim 10 , wherein the dynamic parameter is based on at least one of a mass associated with the autonomous lawn mower or inertia of the autonomous lawn mower.

12. The method of claim 8 , wherein the second current state indicates the current speed and the current rotational rate.

13. A system comprising:

processing circuitry; and

memory communicatively coupled to the processing circuitry and storing executable instructions that, when executed by the processing circuitry, cause the processing circuitry to perform operations comprising:

receiving target data for an autonomous lawn mower, the target data indicating one or more target states;

determining a first difference between a first current state of the autonomous lawn mower and the one or more target states, the first current state including at least one of a current position, a current orientation, a current speed, or a current rotational rate;

based at least in part on the first difference, determining whether to recalculate the target data or provide compensation data to cause the autonomous lawn mower to realign with the one or more target states;

in response to determining to recalculate the target data:

determining new target data for the autonomous lawn mower; and

controlling the autonomous lawn mower based at least in part on the new target data; and

in response to determining to provide compensation data to cause the autonomous lawn mower to realign with the one or more target states:

determining, as the compensation data and based at least in part on the first difference, one or more of a compensation speed or compensation rotational rate;

determining a second difference between a second current state and the compensation data;

determining, based at least in part on the second difference, an amount of force; and

controlling a drive unit of the autonomous lawn mower based at least in part on the amount of force.

14. The system of claim 13 , wherein the operations further comprise:

determining an estimated state of the autonomous lawn mower based at least in part on the amount of force;

determining a subsequent amount of force based at least in part on the estimated state of the autonomous lawn mower; and

controlling the drive unit based at least in part on the subsequent amount of force.

15. The system of claim 13 , wherein the second current state includes the current speed and the current rotational rate.

16. The system of claim 13 , wherein the operations further comprise:

determining, while the autonomous lawn mower is operating, a dynamic parameter related to at least one of a mass associated with the autonomous lawn mower or inertia of the autonomous lawn mower;

wherein determining the amount of force is based at least in part on the dynamic parameter.

17. The system of claim 13 , wherein the one or more target states indicate a target position.

18. The system of claim 13 , wherein the determining the amount of force is based at least in part on a rate of change of a desired state and an output of a dynamic parameter function determined based at least in part on the first current state.

19. The system of claim 13 , wherein the one or more target states indicate at least one of a target position, a target orientation, a target linear velocity, or a target rotational velocity.