Add Boat Mission mode support through rate controllers #20082

Jaeyoung-Lim · 2022-08-21T13:49:06Z

Describe problem solved by this pull request

Previously USVs(boats) did not work in mission mode due to the lack of low level control in the rover position control. Due to the low yaw damping on water the boat model would just oscillate on the yaw axis without meaningful navigation.

Describe your solution

This PR adds a rate control loop on the rover position control module and enables way point navigation for boats.

Added rate control of a rover using the common rate control library
- Added support for Acro mode for rovers (includes Add support for Acro mode on Rovers with Rates Control #18317)
Had to make commander consider boats as a rover vehicle type for waypoint acceptance (Fixes (Asking for Ideas) Exception for Non-aerial vehicle's takeoff altitude check for missions #19045)
This PR needs a modification of the gazebo model for Improved control authority and directional stability of the gazebo boat model Increase control authority of boats PX4-SITL_gazebo-classic#895

Test data / coverage

After PR

Tested in SITL and demonstrated successfully of following a survey pattern

make px4_sitl gazebo_boat

Flight log: Log

Before PR

Flight log: Log

Additional context

This is a extended PR of Add support for Acro mode on Rovers with Rates Control #18317
@junwoo091400 This should be useful for [WIP] Rover controller refactor #19957
Fixes (Asking for Ideas) Exception for Non-aerial vehicle's takeoff altitude check for missions #19045
Depends on Increase control authority of boats PX4-SITL_gazebo-classic#895

Jaeyoung-Lim · 2022-08-22T13:06:36Z

Actuator controls seem to be saturating a lot...

Jaeyoung-Lim · 2022-08-22T15:22:49Z

Retuned rate controller in last commit

jasta · 2022-08-30T17:45:30Z

ROMFS/px4fmu_common/init.d-posix/airframes/1070_boat

+param set-default GND_RATE_D 0.02
+param set-default GND_RATE_FF 0.0
+param set-default GND_RATE_I 0.0
+param set-default GND_RATE_MAX 0.5


Shouldn't this be GND_YAW_RATE for readability? "ground rate" is very vague and could apply to all kinds of things, including being misinterpreted as speed IMO.

@jasta Thanks for the suggestion, and I agree that it would be a better naming. However, by using GND_YAW_RATE, parameters becomes too long (for example GND_YAW_RATE_IMAX ) and is not possible to add as a parameter

jasta · 2022-08-30T17:52:10Z

Thanks for doing this work, I was wondering why my USV was completing missions at such low speeds and it looks like it could be affected by the position controller issues you fixed in this PR. See attached actuator outputs if that's useful to confirm I'm seeing a similar issue:

Note that I also modified the navigator to support RTL mode by disabling the RTL_STATE_CLIMB state, I was hoping to prep a PR myself but got sidetracked trying to understand why the boat executed the RTL mode at such low speeds.

What are we missing with this PR to be able to merge it? Would it be useful if I help field test it?

Jaeyoung-Lim · 2022-08-30T18:33:04Z

@jasta Thanks!

Note that I also modified the navigator to support RTL mode by disabling the RTL_STATE_CLIMB state, I was hoping to prep a PR myself but got sidetracked trying to understand why the boat executed the RTL mode at such low speeds.

This would be awesome, and probably a much needed feature!

What are we missing with this PR to be able to merge it? Would it be useful if I help field test it?

What is missing is a field test on a real boat. I would say this is probably better than the previous state, but it would be great if you could give it a run and see if something critical is missing from this pull request.

Thanks!

jasta · 2022-08-30T19:12:00Z

I'll definitely give this a try, but after some further digging I realized existing parameters may have been the culprit in my case. Default GND_THR_MAX is 0.3f which seems to line up exactly with what the logs are saying. There are several parameters here that are interacting in ways that are very counter intuitive as well, even when looking at the code. For example, GND_SPEED_TRIM doesn't appear to trim the ground speed, but rather sets the target speed if the current setpoint doesn't specify a cruising speed? So if you set a mission cruise speed of 10m/s, but GND_SPEED_TRIM is 3m/s, you'll still get 10m/s? It also only applies if _control_mode.flag_control_position_enabled is true even though RoverPositionControl still may be in charge of setting speed/velocity?

Also, GND_SPEED_THR_SC is for some reason being read by navigator/rtl.cpp? Seems like dead code to me though??? Return mode seems to always publish setpoints with cruising_speed=-1 though, so that should let RoverPositionControl takeover and use appropriate values.

Oh, and I also seemed to have stumbled upon a bug where if you start a mission with a set cruising speed then issue a return command, then resume the mission, the mission speed will be set to cruising_speed=-1 again so you'll get GND_SPEED_TRIM most likely instead. If you change the mission value, re-upload it, then start the mission, this will be once again the mission speed.

Jaeyoung-Lim · 2022-08-30T19:30:42Z

There are several parameters here that are interacting in ways that are very counter intuitive as well, even when looking at the code. For example, GND_SPEED_TRIM doesn't appear to trim the ground speed, but rather sets the target speed if the current setpoint doesn't specify a cruising speed? So if you set a mission cruise speed of 10m/s, but GND_SPEED_TRIM is 3m/s, you'll still get 10m/s? It also only applies if _control_mode.flag_control_position_enabled is true even though RoverPositionControl still may be in charge of setting speed/velocity?

@jasta If you are curious, you can have a look at the logic:

PX4-Autopilot/src/modules/rover_pos_control/RoverPositionControl.cpp

Lines 240 to 259 in e780a58

    
           float mission_target_speed = _param_gndspeed_trim.get(); 
        
           if (PX4_ISFINITE(_pos_sp_triplet.current.cruising_speed) && 
        
               _pos_sp_triplet.current.cruising_speed > 0.1f) { 
        
           	mission_target_speed = _pos_sp_triplet.current.cruising_speed; 
        
           } 
        
           // Velocity in body frame 
        
           const Dcmf R_to_body(Quatf(_vehicle_att.q).inversed()); 
        
           const Vector3f vel = R_to_body * Vector3f(ground_speed(0), ground_speed(1), ground_speed(2)); 
        
           const float x_vel = vel(0); 
        
           const float x_acc = _vehicle_acceleration_sub.get().xyz[0]; 
        
           // Compute airspeed control out and just scale it as a constant 
        
           mission_throttle = _param_throttle_speed_scaler.get() 
        
           		   * pid_calculate(&_speed_ctrl, mission_target_speed, x_vel, x_acc, dt); 
        
           // Constrain throttle between min and max 
        
           mission_throttle = math::constrain(mission_throttle, _param_throttle_min.get(), _param_throttle_max.get());

Although I would try to separate the discussion from this PR since it is not related.

Part of the reason is these things have been staying like this is because rovers were not being really tested/used in auto modes. Most rovers using PX4 were using offboard and direct actuator controls.

Therefore it would be great if we can address these problems one by one as we move forward. If you can address the points of the throttle control in a separate PR or at least post a issue would be helpful.

Also, GND_SPEED_THR_SC is for some reason being read by navigator/rtl.cpp? Seems like dead code to me though??? Return mode seems to always publish setpoints with cruising_speed=-1 though, so that should let RoverPositionControl takeover and use appropriate values.

Oh, and I also seemed to have stumbled upon a bug where if you start a mission with a set cruising speed then issue a return command, then resume the mission, the mission speed will be set to cruising_speed=-1 again so you'll get GND_SPEED_TRIM most likely instead. If you change the mission value, re-upload it, then start the mission, this will be once again the mission speed.

I think these are all related to the RTL use case, which was never really tested with boats and rovers. Therefore we would need to address these problems one by one - also not related to this PR and probably deserves a separate issue/PR

jasta · 2022-08-30T20:04:35Z

Ack'd, thanks for the quick info, I'm going to isolate all of these things and do some more field testing shortly. Hopefully I can provide an isolated report of the efficacy of just this diff once I correct for my baseline numbers with updated parameters.

Jaeyoung-Lim · 2022-09-03T10:32:07Z

@jasta Would you have an ETA for the testing you mentioned? Thanks!

junwoo091400 · 2022-09-03T14:45:23Z

Rebased on main

junwoo091400

Comments on quick points I discovered

junwoo091400 · 2022-09-03T15:27:41Z

src/modules/rover_pos_control/RoverPositionControl.cpp


+	_steering_input = math::constrain(_steering_input + torque(2), -1.0f, 1.0f);


Shouldn't the steering input be set directly by the torque? Why would we accumulate the torque setpoint?

For example when we have achieved our desired rate, the FF gain will then still be accumulating in the steering input, which isn't what we want.

Not really, the angular acceleration maps to the steering rate and not the steering input

How come it's the steering rate? When you want to apply certain torque, shouldn't the rudder deflection (steering) be proportional to that torque setpoint, instead of being integral of it?

If we assume $\dot{\psi} = V * \delta$ where $\delta$ is the steering and $\dot{\psi}$ is the yaw rate.

$\ddot{\psi} = \dot{V} * \delta + V * \dot{\delta}$
We are neglecting the effect of $\dot{V}$ here since we don't want to change $V$ to control the yaw rate. Therefore if we constraint $V$ to be constant,
$\ddot{\psi} \approx V * \dot{\delta}$

How come it's the steering rate? When you want to apply certain torque, shouldn't the rudder deflection (steering) be proportional to that torque setpoint, instead of being integral of it?

If rudder input directly corresponds to torque(or angular acceleration), with a given rudder and speed input you should have the yaw rate accelerating faster and faster. However, it normally converges quickly to a certain yaw rate.

This is assuming that you have sufficient yaw damping, which most rovers have. For boats I am not sure how well this assumption would hold, but guessing that it would be pretty good for differential thrust boats.

If we assume
ψ˙=V∗δ where δ is the steering and
ψ˙ is the yaw rate.

What does V stand for?

If rudder input directly corresponds to torque(or angular acceleration), with a given rudder and speed input you should have the yaw rate accelerating faster and faster. However, it normally converges quickly to a certain yaw rate.

Shouldn't this be handled by the I-term of the PID controller? Or, regarding the damping, is integrating the steering input the common way to mitigate this torque control problem?

What does V stand for?

V stands for velocity

Shouldn't this be handled by the I-term of the PID controller? Or, regarding the damping, is integrating the steering input the common way to mitigate this torque control problem?

I don't think it should be handled by the PID controller assuming steering corresponds to torque (because it does not)

@junwoo091400 If you think using a torque model is the right way, I am all open for it as long as you can show that it is a better way to handle rate control. 😄

For now I have temporarily pushed this single commit to address this change I proposed, but we can revert back, if it proves to be inefficient, after real life testing of the vehicle: 205920e

I think you are probably right, ardurover is also on a pid loop, but with ff https://github.com/ArduPilot/ardupilot/blob/f74d22135cc23b36f142c955314408fb3adc566d/libraries/APM_Control/AR_AttitudeControl.cpp#L572-L573

junwoo091400 · 2022-09-03T15:29:01Z

src/modules/rover_pos_control/RoverPositionControl.cpp

+					///TODO: Max yaw rate
+					float desired_yaw_rate = _gnd_control.nav_lateral_acceleration_demand() / saturated_speed_2d.norm();
+					_rates_sp.roll = 0.0;
+					_rates_sp.pitch = 0.0;
+					_rates_sp.yaw = math::constrain(desired_yaw_rate, -max_yaw_rate, max_yaw_rate);
+					_rates_sp.thrust_body[0] = math::constrain(mission_throttle, 0.0f, 1.0f);
+					_rates_sp.timestamp = hrt_absolute_time();
+					_rates_sp_pub.publish(_rates_sp);
 				}
 			}
 			break;

 		case STOPPING: {
-				_act_controls.control[actuator_controls_s::INDEX_YAW] = 0.0f;
-				_act_controls.control[actuator_controls_s::INDEX_THROTTLE] = 0.0f;
+				_rates_sp.roll = 0.0;
+				_rates_sp.pitch = 0.0;
+				_rates_sp.yaw = 0.0;
+				_rates_sp.thrust_body[0] = 0.0;
+				_rates_sp.timestamp = hrt_absolute_time();
+				_rates_sp_pub.publish(_rates_sp);


We are publishing rate setpoint in 4 different positions in the code. Is this how we want to control / manipulate the _rates_sp variable? Was this unavoidable since the L1 controller's output must be translated into the rate setpoint?

No really, this is mainly because I wanted to add the rate controller first, and then later address the architectural issues in the rover controller in a different PR

Jaeyoung-Lim · 2022-10-06T15:51:36Z

Rebased

altmattr · 2022-10-11T01:05:49Z

I'm in need of this excellent work, but I've hit a bug - can anyone here shed light on it? #20374

Jaeyoung-Lim · 2022-10-16T18:05:12Z

@altmattr You would need this fix in the sitl_gazebo submodule: PX4/PX4-SITL_gazebo-classic#917

altmattr · 2022-10-17T01:46:00Z

@Jaeyoung-Lim Thankyou, that fix has worked for me. I am still unable to get a mission going. Missing upload keeps being rejected due to altitude too low. I'll keep working on it!

Jaeyoung-Lim · 2022-10-17T04:55:52Z

@altmattr Try changing the altitude of your mission?

Jaeyoung-Lim · 2023-08-02T20:30:21Z

@junwoo091400 I am against merging this without proper testing

slgrobotics · 2023-08-02T22:37:42Z

Just to clarify - I brought up this PR to attention of the Team because I see the value of having a vehicle yaw controlled by the best generic mechanism available - RateControl module.

IMHO this is applicable to all ground vehicles, but there should be three distinct modules:

An example small robotic Rover, like Avion R1. All that's needed here is a PID for speed/thrust control and a RateControl module for yaw/torque control - basically, just small touches to the existing RoverPositionControl
Boat control module - to extend the above with boat-specific features.
Large Rovers. In my case, I had not only to control speed and yaw as above, but divide vehicle path into phases - Turn, Departure, Line Following, Arrival, Stopping - and use a State Machine for that. It is working well for my gas engine driven zero turn mower, and might work for Ackermann-steered vehicles with significant inertia.

I don't have a boat, but I do have a mid-sized differential drive robot (14" wheels) to test case #1, and could probably have R1 rover fixed in classic Gazebo simulation.

My code is here, just in case: https://github.com/slgrobotics/PX4-Autopilot/tree/main/src/modules/rover_pos_control

Jaeyoung-Lim · 2023-08-02T23:25:09Z

@slgrobotics Thanks for the context.

As you mentioned, this can also be tested on a rover. What I would like to see is what reasonable tuning values would be, and that it working properly without any major problems.

However, testing would need to be preceded before we get this PR in. I wouldn't feel comfortable merging something that is not tested and potentially promising.

not7cd · 2023-08-03T08:53:56Z

What would suffice as proper testing? My team might assist with that. We have an ASV and it was already running code from this PR during RoboBoat. But we didn't fully test PID parameters.

Jaeyoung-Lim · 2023-08-03T09:11:19Z

@not7cd

We have an ASV and it was already running code from this PR during RoboBoat.

Ha, interesting!

Great, a flight log from the following tests would fit into sufficient testing.

Drive in acro mode with step input on the yaw axis to see that the vehicle correctly stabilizes and maintains a constant yaw rate
Same test as above but on different speed to see that the PID adapts correctly to different speeds
Demonstration in mission mode where the vehicle tracks different waypoints

Jaeyoung-Lim · 2023-08-09T07:47:58Z

@not7cd @slgrobotics Any updates?

slgrobotics · 2023-08-10T00:42:46Z

@Jaeyoung-Lim - I've got my rover working just now, will be experimenting with your code tomorrow. Sorry for delay.

slgrobotics · 2023-08-11T16:22:44Z

@Jaeyoung-Lim - It's a bit rainy here in Alabama and my progress is slow.

Here is what I have so far:

My Rover requires some modifications, for which I have a fork from PX4 "main". To test this PR I copied its rover_pos_control folder over to mine. As this PR's origin is way outdated, it requires modifications to compile (see patch below). It makes sense to bring PR code to current "main" state.
I am surprised to see GND_WHEEL_BASE gone, but looking at the code I see the reason. Will keep an eye on that.
When I run the robot in Manual mode, it seems to be configured so that it doesn't move backwards, and stops only when I pull the stick all the way down. No big deal, but worth looking into that. Rovers and boats can be put in reverse.
So far I could only test my Rover in a short mission. I had to set D=I=FF=0 and P=0.5 to avoid oscillations, that needs to be tuned. The robot seems to roughly move to a waypoint, but doesn't proceed to the next one. Obviously, lots of tuning to follow.

Here is what I did to make PR's code compile with current "main":

diff /home/sergei/src20/PX4-Autopilot/src/modules/rover_pos_control/./RoverPositionControl.cpp src/modules/rover_pos_control/./RoverPositionControl.cpp
96c96
< 		_gnd_control.set_l1_roll_limit(math::radians(0.0f));
---
> 		//_gnd_control.set_l1_roll_limit(math::radians(0.0f));
105c105
< 		_rate_control.setGains(matrix::Vector3f(0.0, 0.0, _param_rate_p.get()), matrix::Vector3f(0.0, 0.0, _param_rate_i.get()),
---
> 		_rate_control.setPidGains(matrix::Vector3f(0.0, 0.0, _param_rate_p.get()), matrix::Vector3f(0.0, 0.0, _param_rate_i.get()),
170,171d169
< 					_act_controls.control[actuator_controls_s::INDEX_ROLL] = 0.0f; // Nominally roll: _manual_control_setpoint.roll;
< 					_act_controls.control[actuator_controls_s::INDEX_PITCH] = 0.0f; // Nominally pitch: -_manual_control_setpoint.pitch;
173c171
< 					_act_controls.control[actuator_controls_s::INDEX_YAW] =
---
> 					_torque_control =
176c174
< 					_act_controls.control[actuator_controls_s::INDEX_THROTTLE] = (_manual_control_setpoint.throttle + 1.f) * .5f;
---
> 					_thrust_control = (_manual_control_setpoint.throttle + 1.f) * .5f;
384c382
< 		_act_controls.control[actuator_controls_s::INDEX_THROTTLE] = math::constrain(control_throttle, 0.0f, mission_throttle);
---
> 		_thrust_control = math::constrain(control_throttle, 0.0f, mission_throttle);
400c398
< 		_act_controls.control[actuator_controls_s::INDEX_YAW] = control_effort;
---
> 		_torque_control = control_effort;
404,405c402,403
< 		_act_controls.control[actuator_controls_s::INDEX_THROTTLE] = 0.0f;
< 		_act_controls.control[actuator_controls_s::INDEX_YAW] = 0.0f;
---
> 		_thrust_control = 0.0f;
> 		_torque_control = 0.0f;
447c445
< 	_act_controls.control[actuator_controls_s::INDEX_YAW] = _steering_input;
---
> 	_torque_control = _steering_input;
451c449
< 	_act_controls.control[actuator_controls_s::INDEX_THROTTLE] =  math::constrain(control_throttle, 0.0f, 1.0f);
---
> 	_thrust_control =  math::constrain(control_throttle, 0.0f, 1.0f);
568,569c566,567
< 			_act_controls.timestamp = hrt_absolute_time();
< 			_actuator_controls_pub.publish(_act_controls);
---
> 			//_act_controls.timestamp = hrt_absolute_time();
> 			//_actuator_controls_pub.publish(_act_controls);
573c571
< 			v_thrust_sp.xyz[0] = _act_controls.control[actuator_controls_s::INDEX_THROTTLE];
---
> 			v_thrust_sp.xyz[0] = _thrust_control;
580,582c578,580
< 			v_torque_sp.xyz[0] = _act_controls.control[actuator_controls_s::INDEX_ROLL];
< 			v_torque_sp.xyz[1] = _act_controls.control[actuator_controls_s::INDEX_PITCH];
< 			v_torque_sp.xyz[2] = _act_controls.control[actuator_controls_s::INDEX_YAW];
---
> 			v_torque_sp.xyz[0] = 0.0f;
> 			v_torque_sp.xyz[1] = 0.0f;
> 			v_torque_sp.xyz[2] = _torque_control;
diff /home/sergei/src20/PX4-Autopilot/src/modules/rover_pos_control/./RoverPositionControl.hpp src/modules/rover_pos_control/./RoverPositionControl.hpp
78d77
< #include <uORB/topics/actuator_controls.h>
115d113
< 	uORB::Publication<actuator_controls_s>		_actuator_controls_pub{ORB_ID(actuator_controls_0)};  /**< actuator controls publication */
133d130
< 	actuator_controls_s				_act_controls{};		/**< direct control of actuators */
161c158,160
< 	float _steering_input{0.0};
---
> 	float _steering_input{0.0f};
> 	float _torque_control{0.0f};
> 	float _thrust_control{0.0f};

slgrobotics · 2023-08-14T16:55:02Z

@Jaeyoung-Lim - OK, I tried to run my rover under the PR code (adapted to current PX4 "main" state) and it didn't go well. I don't see any problem with the code itself - the rate controller worked fine, but the EKF2 heading was totally messed with my sensors (mpu9250+hmc5883) and the robot couldn't hold lines or follow waypoints. Things work better in sim, when I use "make gazebo-classic_r1_rover" build but replace R1 model with my own. EKF2has always been a problem for me, and on my big lawnmower I just use dual RTK GPS for heading directly to avoid such issues.

At the moment I can't be much help (although I might keep an eye on this) - so I hope there will be somebody else for testing.
My suggestion is to bring this PR up to date with "main" and keep trying.

BTW, the reverse not working issue will go away after line 148 is merged with "main" (no need for +1.f and *0.5f).

junwoo091400 · 2023-08-14T16:57:30Z

robot couldn't hold lines or follow waypoints

Do you have a log to share? That would be helpful!

slgrobotics · 2023-08-15T02:53:13Z

@Jaeyoung-Lim - I have log files, not sure if they are of any use. Normally I would make notes about conditions, reasoning and parameters, but these runs were very unsatisfactory - so I didn't get to that stage. I didn't really try to tune up the PIDs. Default P=5 caused gross oscillations.

https://review.px4.io/plot_app?log=ea472c57-9097-4034-930d-8d367850bcf7 (P=0.5)

https://review.px4.io/plot_app?log=b75f9566-43cc-452f-8d3b-5d3ef1543776 (P=0.1)

https://review.px4.io/plot_app?log=cbfe2842-8c9f-4487-b6e5-282722b17015

https://review.px4.io/plot_app?log=ed37a0a8-c7b2-41d1-beca-e4fd60865f87

I can run the beast again if you feel that would help.

This commit adds the rover rate controller for rover

Rebase fixes

slgrobotics · 2023-08-15T15:43:37Z

@Jaeyoung-Lim - I tried to build the latest:
RoverPositionControl.cpp : 102 - should be setPidGains(...)
431 - should be _yaw_control = ... (act_controls deprecated)
435 - should be _throttle_control = ... - and maybe constrain to -1.0f +1.0f ?
builds ok after corrections above.

slgrobotics · 2023-08-16T00:50:10Z

@Jaeyoung-Lim - I ran your latest code on my rover. This time I had a chance to play with parameters, and EKF2 was behaving reasonably well. My biggest problem was that with the speed controller delivering thrust values close to 1.0 the torque (yaw) values were added and actuators (wheels) were basically saturating. Robot refused to turn. Multicopter controller senses actuators saturation, and that could be done for diff drive rovers, if needed - but that's beyond the scope of this PR. I just trimmed the speed setpoint to a crawl to have it do the mission.

Here is the latest log - there is a lot more to tune though:

https://review.px4.io/plot_app?log=84ce4592-a79e-40ab-b9b3-dfdb3816d9eb

BTW, GND_ATT_P isn't used anymore

DronecodeBot · 2023-10-27T15:56:49Z

This pull request has been mentioned on Discussion Forum for PX4, Pixhawk, QGroundControl, MAVSDK, MAVLink. There might be relevant details there:

https://discuss.px4.io/t/vectored-thrust-for-surface-vessel-i/34920/2

DronecodeBot · 2024-01-26T08:23:14Z

This pull request has been mentioned on Discussion Forum for PX4, Pixhawk, QGroundControl, MAVSDK, MAVLink. There might be relevant details there:

https://discuss.px4.io/t/rc-speed-boat-with-px4-episode-1-using-px4-to-control-the-boat/28429/10

hamishwillee · 2024-01-30T23:49:26Z

Given this derives from rovers, would I be correct in saying Rovers don't support mission mode? What modes are supported? Just "manual" and "offboard"? Need to update https://docs.px4.io/main/en/frames_rover/ to say this.

Jaeyoung-Lim · 2024-01-31T12:48:43Z

@hamishwillee Not really, when I was writing this, rover mission worked but failed on boars due to the low damping over water

PyroX2 · 2024-02-10T12:49:56Z

@Jaeyoung-Lim Hi, we are using rover position control to control our custom ASV. We are currently at roboboat and we are trying to set waypoints based on position of buoys that camera sees. It seems to work on the ground but when we test it on water our ASV sometimes it swims to the side when it should forward. Could you suggest based on this log file if it is a problem with our ROS2 code or pixhawk configuration? Roboboat ends on sunday so we would be really grateful for any quick response. Thanks in advance
Log file: https://logs.px4.io/plot_app?log=991b55fe-7fea-471b-b77a-2353e4648201

cc @not7cd

Jaeyoung-Lim · 2024-02-10T12:53:55Z

@Kubol307 It is hard to know just looking at the log what your expectation is and how it is i plemented.

Have you tested the rate controller without your ros2 code? You need to isolate the two components so that you can identify where your problem actually is

PyroX2 · 2024-02-10T14:12:21Z

@Jaeyoung-Lim We have just tested it on water and it's most likely a problem with our ROS2 code. When we are setting waypoints from QGC everything worked fine. However what's interesting is that when we tested it now everything worked fine until it started to swim in circles no matter what waypoints we send to it. Pixhawk reboot helped but it's not the first time it happened. The next time it occurs I will check if it's the same with waypoint from QGC. Here is the log file from this run: https://logs.px4.io/plot_app?log=e4226fff-58e4-4354-b27f-ab4b48fd8b33
And here spinning began:

PyroX2 · 2024-02-10T14:14:09Z

And also our PID gains. Earlier, with higher P on rate controller we experienced really unstable movement. Setting it to about 0.05 helped a lot.

Jaeyoung-Lim · 2024-02-10T14:14:21Z

@Kubol307 This PR implements a yaw rate controller. You need to check if the yaw rate controller is well tuned before analyzing position errors

PyroX2 · 2024-02-10T21:51:47Z

@Jaeyoung-Lim What tests would help you to merge this pull request?

Jaeyoung-Lim · 2024-02-11T08:20:24Z

@Kubol307 I am not sure how would merging this PR help you in the roboat competition?

If you want to hop on a call, I will be happy to help. You can reach out to me on the PX4 discord

PyroX2 · 2024-02-11T21:33:17Z

@Jaeyoung-Lim Unfortunately we have already had our last run on roboboat. We might have found what was the problem as we found this issue: https://discuss.px4.io/t/rover-waypoint-altitude-issue/7455 Before we initialized pixhawk on the ground and then put the boat on water. Today we tried to push it a little into the water so that pixhawk initializes while being in lowest point and we didnt experienced the above issue. May the altitude actually be the problem? If you need any additional information I will be happy to provide you with that

Jaeyoung-Lim requested review from bresch, junwoo091400 and julianoes August 21, 2022 13:49

Jaeyoung-Lim added Rover 🚙 Rovers and other UGV Boat 🚤 labels Aug 21, 2022

Jaeyoung-Lim marked this pull request as ready for review August 21, 2022 14:02

Jaeyoung-Lim changed the title ~~Boat Mission mode support through rate controllers~~ Add Boat Mission mode support through rate controllers Aug 21, 2022

Jaeyoung-Lim mentioned this pull request Aug 21, 2022

Add support for Acro mode on Rovers with Rates Control #18317

Closed

Jaeyoung-Lim requested a review from dagar August 21, 2022 16:07

This was referenced Aug 21, 2022

(Asking for Ideas) Exception for Non-aerial vehicle's takeoff altitude check for missions #19045

Closed

Increase control authority of boats PX4/PX4-SITL_gazebo-classic#895

Merged

Jaeyoung-Lim force-pushed the pr-rover-ratesp branch from 4e27d00 to 2afd02b Compare August 22, 2022 15:31

jasta reviewed Aug 30, 2022

View reviewed changes

junwoo091400 force-pushed the pr-rover-ratesp branch from 2afd02b to 89b093b Compare September 3, 2022 14:44

junwoo091400 reviewed Sep 3, 2022

View reviewed changes

Jaeyoung-Lim force-pushed the pr-rover-ratesp branch from 89b093b to d2ffc34 Compare October 6, 2022 15:51

junwoo091400 added the Maintainers call Add items that should be discussed in the weekly maintainers call! label Aug 2, 2023

Jaeyoung-Lim added 3 commits August 15, 2023 14:39

Add rate controller for rover

840d76e

This commit adds the rover rate controller for rover

Do not integrate steering input

62f0faf

update angular accelerations

aa004b9

Rebase fixes

Jaeyoung-Lim force-pushed the pr-rover-ratesp branch from 193858d to aa004b9 Compare August 15, 2023 12:39


		_steering_input = math::constrain(_steering_input + torque(2), -1.0f, 1.0f);

Add Boat Mission mode support through rate controllers #20082

Are you sure you want to change the base?

Add Boat Mission mode support through rate controllers #20082

Conversation

Jaeyoung-Lim commented Aug 21, 2022 • edited Loading

Describe problem solved by this pull request

Describe your solution

Test data / coverage

After PR

Before PR

Additional context

Jaeyoung-Lim commented Aug 22, 2022

Jaeyoung-Lim commented Aug 22, 2022 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jasta commented Aug 30, 2022 • edited Loading

Jaeyoung-Lim commented Aug 30, 2022 • edited Loading

jasta commented Aug 30, 2022

Jaeyoung-Lim commented Aug 30, 2022 • edited Loading

jasta commented Aug 30, 2022

Jaeyoung-Lim commented Sep 3, 2022

junwoo091400 commented Sep 3, 2022

junwoo091400 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Jaeyoung-Lim Sep 3, 2022 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Jaeyoung-Lim Sep 10, 2022 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Jaeyoung-Lim Sep 3, 2022 • edited Loading

Choose a reason for hiding this comment

Jaeyoung-Lim commented Oct 6, 2022

altmattr commented Oct 11, 2022

Jaeyoung-Lim commented Oct 16, 2022

altmattr commented Oct 17, 2022

Jaeyoung-Lim commented Oct 17, 2022

Jaeyoung-Lim commented Aug 2, 2023

slgrobotics commented Aug 2, 2023

Jaeyoung-Lim commented Aug 2, 2023

not7cd commented Aug 3, 2023

Jaeyoung-Lim commented Aug 3, 2023 • edited Loading

Jaeyoung-Lim commented Aug 9, 2023

slgrobotics commented Aug 10, 2023

slgrobotics commented Aug 11, 2023

slgrobotics commented Aug 14, 2023 • edited Loading

junwoo091400 commented Aug 14, 2023

slgrobotics commented Aug 15, 2023

slgrobotics commented Aug 15, 2023

slgrobotics commented Aug 16, 2023 • edited Loading

DronecodeBot commented Oct 27, 2023

DronecodeBot commented Jan 26, 2024

hamishwillee commented Jan 30, 2024

Jaeyoung-Lim commented Jan 31, 2024

PyroX2 commented Feb 10, 2024

Jaeyoung-Lim commented Feb 10, 2024 • edited Loading

PyroX2 commented Feb 10, 2024

PyroX2 commented Feb 10, 2024

Jaeyoung-Lim commented Feb 10, 2024

PyroX2 commented Feb 10, 2024

Jaeyoung-Lim commented Feb 11, 2024

PyroX2 commented Feb 11, 2024

Jaeyoung-Lim commented Aug 21, 2022 •

edited

Loading

Jaeyoung-Lim commented Aug 22, 2022 •

edited

Loading

jasta commented Aug 30, 2022 •

edited

Loading

Jaeyoung-Lim commented Aug 30, 2022 •

edited

Loading

Jaeyoung-Lim commented Aug 30, 2022 •

edited

Loading

Jaeyoung-Lim Sep 3, 2022 •

edited

Loading

Jaeyoung-Lim Sep 10, 2022 •

edited

Loading

Jaeyoung-Lim Sep 3, 2022 •

edited

Loading

Jaeyoung-Lim commented Aug 3, 2023 •

edited

Loading

slgrobotics commented Aug 14, 2023 •

edited

Loading

slgrobotics commented Aug 16, 2023 •

edited

Loading

Jaeyoung-Lim commented Feb 10, 2024 •

edited

Loading