Designing a magical software that helps automotive companies make smarter and safer cars internationally

Overview

As the sole Product Designer, I initiated the design and architecture of an Advanced Driver Assistance System ​(ADAS) editor, focusing on developing a cohesive, intuitive editor for users to author scenarios.

Successfully enhanced scenario creation improving user satisfaction and aligning the studio’s features into our customer needs. The studio’s launch was successful by the increase of 54% surge in orders, leading to more feature requests pioneering us to be one of the most reliable products in the market.

53%

Increase of orders

11%

Boost in revenue

Contracts with large automotive companies

Overview of company progression

I joined RightHook in 2022 when they were developing WorldSim for their reccurring customer Honda. I was a part of the design efforts in making WorldSim a sustainable desktop application which allowed RightHook to extend their contract with Honda and eventually lead to their acquisition by driving simulator company VI-Grade.

So, what does this vehicle software exactly do?

Think of it this way, when a vehicle is being manufactured, it needs to be tested in a realistic environment under different conditions. For ADAS and Autonomous Vehicle Development, automotive companies and suppliers mainly focus on these key area:
✔️ The vehicle must be able to detect its surroundings, including other vehicles, pedestrians, sometimes animals, road work and other unexpected obstacles.

✔️ All information collected by sensors must be combined, checked and used as inputs for all control algorithms that are in charge to predict the behavior of autonomous or assisted cars.


At the time car companies were using real world staged environments to test their vehicle sensors. This often gave limited testing results and very expensive. See example below.

This is where VI-WorldSim comes into play, our software was able to construct an infinite amount of simulation data with various minute changes that can be tested directly on vehicle sensors without having an actual real world environment. Saving time and money.

Through user interviews and extensive discussions with the CEO and CTO, we defined our objective as researching and designing their desktop application to have a more human friendly approach where data scientists not only have the opportunity to define the scenarios they are creating but also visualize them in a 3D viewport.

Tracking KPIs

When we were supplying one manufacturer (Honda), retention was easy to track because we would have recurring conversations with their team showing them our updates to the software as well as extending their contract to prioritize their requests.

However after the acquisition of RightHook by VI-Grade we took on their existing customer base Audi, Volvo and Ferrari which led to a redesign of the software to make it more sustainable. With the quick redesign we then added customer acquisition to our main focus acquiring contracts with Ford, Goodyear, Aston Martin and multiple other manufacturers world wide.

Reviewing the product at the beginning stage

When I joined the product was not what it is today. The software had some fundamental features the CTO thought was enough to create a simple scenario. So, conducting a ux review to identify usability issues and capabilities was the best method to move forward while keeping these goals in mind:

1. Analyze the hierarchy of the panels
2. How intuitive is it for new users to understand and navigate through out the studio?
3. Buttons and interactions
4. Are there common workflows?

What was the result?

The UX audit allowed me to discover the softwares key pain point:

  • Lacking a universal hierarchy structure for new users

  • Lacking established components with respective interactions

  • Lacking fundamental visual design; iconography, typography and usage of darker colors.

  • Lacking additional editing features.

Competitive Analysis

I was new to the automotive industry, learning more about our current competitors however at the time the market was very niche. Facing challenges of extremely limited data, I took it upon myself to broaden my research to 3D softwares and simulation games.

What was the result?


I understood the comprehensive understanding of industry standard interface, technology limitations, user navigation in 3D environments and the unique challenges presented by softwares that incorporates both 2D and 3D components.

Designing a beta version of the software with limited research and tight deadline.

With this first design update to the software we were able to make amazing strides to standardizing a design system and making a more cohesive layout.

However updating a full on software didn’t happen without its hurdles. As RightHook already had one client we had to weigh out what’s best for the business vs what’s best for the client. In that same vein we also only had our one client to gather feedback so it was a bit of an echo chamber.

Rewind!

With all the progress to the software steadily in motion and the waves the company was making in the industry we were able to get acquired 🎉. So before we could gather our results on the first redesign we were hit with a flurry of new users and requests to make a full release of our software. This led to our second redesign where I mainly wanted to focus on expanding our sustainability efforts.

Prioritizing testing goals

I held a discussion with the CEO and CTO to create a list of items we wanted to prioritize during testing. We ruled the goals for the research should be to:
  • Understand customer pain points when creating scenarios and navigating through a 3d space.
  • Identify what users might be missing when creating their scenarios.
  • Determine what features are under utilized and if there’s the users priority.
  • Discover opportunities to improve the design.
  • Learn about the users behavior and preferences.

Moderated usability testing

I tested on 7 participants having them run the studio and try to complete a few tasks focusing on:
  • Scenario creation of a simple scene
  • Descriptive scenario creation with instructions

Affinity diagram

At this point I have gathered a great amount of data from usability testing and user interviews featuring user opinions, needs, observations etc. I needed to format everything into a clear space so I decided to create an affinity diagram.

I grouped together similar ideas and created categories to organize user feedback. Then I brought in the CEO and CTO to rank the clusters in order of importances based off the companies priorities.
  • What are some features we have in the studio that are broken?
  • What are some features we can improve on?
  • Future features to improve the software
  • User-behavior Suggestions

Key insights and findings

  • Agent placement: Users are needing to place down agents quicker Users find themselves spending more time adjusting properties of vehicle after placement
  • Navigation: Users get lost in larger maps 80% of users are in a birds eye view when placing down agents
  • Triggers: Users find themselves needing a more expansive trigger system Some of the UI isn’t clear to users to understand
  • Hierarchy: Users find themselves wanting to see connects between agents and script components

Solution: Fresh look on the layout

As the number of customers increased, we began receiving requests for new content and features. We consulted to strategize the optimal time to reorganize the studio, ensuring adequate space for upcoming features and addressing issues associated with the current implementations.

Goal: Re-structure the studio to introduce new and upcoming functions, while allocating appropriate time from the engineers and on my end to not end up with a mess of bugs.

Result: Introduced a file menu to serve as a central location for managing the scenarios and settings, and moved the toolbar to the left side to create a more breathable working space.

Solution: Agent Placement

After analyzing the research, the initial design was cumbersome and occupied excessive space in the viewport. Users had to adjust vehicle behavior post-placement, turning it into a more cumbersome task.
Goal: Create a compact version that allows users to modify actors during placement mode.

Result: Provided a more streamlined experience for placing multiple types of actors, ultimately enhancing user satisfaction and efficiency.

The workflow of creating a sceanrio

  1. Starting off by selecting an environment to start their scenario
  1. Select the actors which consist of; vehicles, pedestrians, animals and obstacles.
  1. Make adjustments from here, users can relocate their actors and adjust their properties accordingly.
  1. Add script components to make the scenario lively.

Maintenance | Design System

As the product expanded, the necessity for a design system arose to maintain components and facilitate swift wireframe design in the ideation phase.

Goal: Establish a unified design system midway through the process.

Result: Created an evolving design system with the core components to help myself and the technical artists to grab assets needed during implementation.

3D interactions | Desiging around 3D spaces

This is a fun one! I didn't just design 2D panels for the software; I also played a role in enhancing the user experience with 3D objects. Below are some of the implemented interactions in the studio.